When you upload files to GitHub, you’ll store them in a “Git repository.” This means that when you make changes (or “commits”) to your files in GitHub, Git will automatically start to track and manage your changes. There are plenty of Git-related actions that you can complete on GitHub directly in your browser, such as creating a Git repository, creating branches, and uploading and editing files.

Once you start to collaborate with others and all need to work on the same repository at the same time, you’ll continually:

• Pull/fetch all the latest changes made by your collaborators from the remote repository on GitHub.
• Push back your own changes to the same remote repository on GitHub.

Git figures out how to intelligently merge this flow of changes, and GitHub helps you manage the flow through features such as “pull requests.”

Environment Setup

In the following section, you will get to see the step-by-step process of setting up VS Code and connecting to a Git repository.

Install Visual Studio Code

You can download VS Code exe file on your PC using the following link. Use the downloaded file to install VS Code.
Link: https://code.visualstudio.com/download

Install Git

Use the following link to download the latest Git file on to your PC. Once the exe file is downloaded, proceed to install task.
Link: Git – Install for Windows

Check the Git installation

On VS Code, navigate to View > Terminal and the following window appears. Enter the following command to get the current installed version information of Git.

git --version

If you want to know the exact path of installation, you can enter the following command.

where git.exe

Create a working directory using VS Code

It is a good practice to maintain a separate folder for each project. To open a folder using VS Code, you can either navigate to the Explorer icon on the left and click on Open Folder or Open a folder from the File menu. Select the existing folder or create a new one and open it by clicking on the “Select folder” button as shown below. This will be your working directory for your project files.

Initialize a local Git repository

Once a new folder is open, it will not have source control enabled for the files within it. In order to enable source control features, you need to initialize a Git repository. Navigate to Source Control > Initialize Repository.

Once you click on Initialize Repository, the source control features will be enabled and the next screen will appear as follows.

The other way of initializing a Git repository is through command. Run Git init in the directory. This creates a hidden .git subdirectory that contains all the necessary metadata and internal databases for Git to start tracking changes. In some cases, we need to reinitialize a Git repository, during which the message will show up as Reinitializing existing Git repository after the Git init command has been entered.

Staging a file

Git add command is used to place changes into the staging area (also known as the “index”), which acts as a temporary holding area for the next commit. While git add . as seen in the above screenshot means to add all changes in the current directory and its subdirectories to the staging area. We will go into the details of commit in the subsequent sections.

Git status command is used to display the current state of your working directory and staging area, showing which files are staged, modified, or untracked.

Connecting to remote repository

Now that the files are staged, next is to connect to the remote repository. The git remote add origin command is used to link a remote repository to your local Git repository. This allows you to push and pull changes between your local and remote repositories. The origin here refers to the remote repository URL. Before this, please create your account on GitHub and create a new repository.

The git remote -v command is used to list existing remote repos along with their fetch and push URLs.

Committing a file

Once the file/s are staged, the next step would be to commit the files. The commit command saves a snapshot of your staged changes to the local repository. Committed snapshots can be thought of as “safe” versions of a project.

The git commit -m "initial commit" is the most common way to commit ,i.e, with a message. This saves staged changes and allows you to include a descriptive message directly in the command line.

Changing your remote repository

If you are in one repository and want to change to another remote repository, you can specify the below command. The same is highlighted in red frame in the above screenshot.

git remote set-url origin https://github.com/sampathgithub1/Test

In the above command, you are specifying the destination remote repository that you want to make use of. Create the “Test” remote repository manually on GitHub using the GUI.

Uploading to remote repository

git push <remote> <branch name> command takes in two arguments. In the above screenshot, you can find the git push origin main command that refers to pushing your local main branch changes to your online repository (origin, an alias). Your remote Git repository will appear as follows on Github after uploading is complete. Login to your Github account using the URL Link and navigate to your repository.

In case you make changes to the source file, you need to perform a commit again using the following command.

Git commit -m "second change commit"

This should be followed by the git push command again. In the above screenshot you can observe the “second change commit” text mentioned next to the source file.

Further Reading:

• An Introduction to OCI Cloud Operations
• AI-Powered ITSM with n8n: Integrating Jira SM, OpenAI and Slack
• An Introduction to AI-Powered Software Testing
• Exploring LLMs and OCI Generative AI
• Data Analysis using Python

Software companies aim to deliver a high-quality product on time and under budget. This is where software testing becomes imperative to deliver a quality product. Software Testing is a systematic process of evaluating an application to identify defects, bugs, or deviations from the expected behaviour. It is fundamentally a process of verification and validation. For valid results, tests must be performed in environments that closely resemble the production environment, using similar infrastructure and data. High-quality products are never an accident; they are the result of deliberate, intelligent effort in development and testing.

Software Testing & STLC

Similar to the SDLC for creating software, there is STLC for Software testing, which needs to be followed by testing teams to ensure software quality. Skipping the testing phase can negatively impact both the product and the brand’s reputation. The six phases of STLC are: Requirement analysis, Test planning, Test case development, Environment setup, Test execution, Test cycle closure. The testing process is categorized into three primary levels as follows, moving from granular components to the entire system.

• Unit Testing
• Integration Testing
• System Testing/E2E test

Functional testing focuses on what the system does (features wrt requirements), while non-functional testing focuses on how it performs (speed, reliability and security). We have the Unit testing, Integration testing, System testing(E2E test), Interface testing, Regression testing and UAT under functional testing. While the documentation testing, installation testing, performance testing, reliability testing, security testing fall under non-functional testing.

Manual or Automation testing is utilized for the above based on strategic purpose. While manual testing is used in the areas where exploratory, human judgement and subjective human experiences are needed. On the other hand, automation testing is used where speed, reliability is needed. Even for repetitive tasks and complex end to end workflows Automation testing is used.

 AI-Powered Testing

AI can act as a Co-pilot to the tester in the following areas of STLC. This can be for Test case generation, Test automation scripts, Test data generation, Test maintenance, prioritisation, and exploratory testing ideas. It is important to note that blind copy-paste of AI output without human verification is dangerous. AI generated test cases can sometimes be broad but lack the necessary depth.

Playwright Framework & Tools

Playwright is a powerful and versatile automation library developed by Microsoft. It enables developers and testers to automate web applications across multiple browsers with ease. Playwright’s architecture is designed to provide a robust, flexible, and high-performance framework for browser automation. It is preferred over older frameworks like Selenium for writing end-to-end test automation scripts.

If you are using a Windows PC, select the necessary option/s and download the latest Node.js using the following link, which is a prerequisite for installing Playwright. Even npm installation is a prerequisite.

Download nodejs

Once the download and installation are complete, use the following commands to check whether it is installed properly or not. The commands can be used in the command prompt or in the Terminal section of VS Code IDE to verify the versions installed. Here, “npm” refers to Node Package Manager.

If you need to download and setup VSCode, use the following link to download and perform the installation.

Download VSCode

On VS Code, click on “Open Folder” to navigate to an existing folder or create a new one (ECOMWEBTEST) and Open it. We will now go through the following link for more details on playwright setup.

Installation | Playwright

As mentioned in the above link, the following command adds Playwright to the existing project/folder on VS Code if it is not yet setup. Once the following command is executed, you can see the folders/files as displayed in the below picture to get populated under the folder ECOMWEBTEST.

npm init playwright@latest

Test Case Generation

Now that we have set up Playwright in VS Code, let us use an AI tool (OpenAI/Claude etc) to generate Test cases for a sample scenario using the following prompt.

"Consider yourself as a QA testing expert. Generate detailed test cases for saucedemo.com login and checkout flow. Include positive and negative scenarios. I will be using playwright in order to test this."

After all the test cases are generated using the above prompt, we will move on to the execution of the test cases.

Test Case Execution

As you are inside the project folder on VS Code, you can refer to the below prompt result to gain insights on how to run the tests.

"I am inside a folder in VS Code. Guide me on how to run the tests listed above."

From the result of the prompt in the earlier section, copy and paste any one of the test cases into the code file. For example, let us copy the code for login test into the newly created code file under tests folder. The name of the file is set to saucedemo.spec.ts as shown in the below picture.

The code to test the login (generated by AI tool) is as follows. Check the code to confirm whether the website name is mentioned or not.

import { test, expect } from '@playwright/test';
test('Successful login with valid credentials', async ({ page }) => {
  await page.goto('https:www.saucedemo.com/');
  await page.fill('[data-test="username"]', 'standard_user');
  await page.fill('[data-test="password"]', 'secret_sauce');
  await page.click('[data-test="login-button"]');
  await expect(page).toHaveURL(/inventory.html/);
  await expect(page.locator('.inventory_item')).toHaveCount(6);
});

Execute the above code by using the following command.

npx playwright test tests/saucedemo.spec.ts --headed

Headed Mode enables testers to visually verify automated actions, such as typing and clicking. Whereas the other mode, i.e, Trace Viewer provides a step-by-step trace of every action for verification. Once you press the enter key, the code will automatically run and execute the login for multiple browsers (Chrome, Webkit and Firefox). On successful run, you will get the following output under the Terminal section.

In case the test is a success as shown above, move on to the next test case. Similar to the first test case, execute the code for other tests using different file names.

Further Reading:

• Maximize Your Potential : The AI Tools You Need to Know About
• Exploring LLMs and OCI Generative AI
• Agentic AI for Agile Teams: Smarter Reporting and Insights
• AI-Powered ITSM with n8n: Integrating Jira SM, OpenAI and Slack
• Navigating Jira : From Basics to Advanced

Space/Page/Page tree

Your content lives in pages ,i.e, the documents you create on your Confluence site. You can create pages for almost anything, from project plans to meeting notes, troubleshooting guides, SOPs, policies, and more. Use the Presenter mode option from More actions(…) of a published page to present it in full screen. Pages are stored in spaces,i.e, the workspaces where you can collaborate on work and keep all your content organized. It’s best to group related content together in the same space, but you can create as many or as few spaces as your team needs. Organize space content with a hierarchical page tree that makes finding work quick and easy. The page history helps you understand the number of changes made by users within a specific period of time. In addition, you can create a completely centralized content repository by hyperlinking the various pages.

Macros

Using macros helps you to extend the capabilities of your Confluence pages and live docs, allowing you to add extra functionality or include dynamic content. Typing “/” opens a list of available macros, allowing you to quickly search and insert them into a page without navigating through the toolbar. I will list a few of the macros that I have come across as follows.

• Table of contents
  • An automated table of contents can be built into Confluence pages. The macro automatically picks up the contents based on the headings within a page.
• Status
  • The Status macro displays a colored rounded box that is useful for reporting the status of a page, project, or task. It’s a simple yet effective way to communicate progress, priority, or any other relevant status information. 
• Roadmap planner
  • Add the Roadmap Planner macro to a page to create a simple, visual timeline that’s useful for planning projects, releases and much more. It’s a simple yet effective way to represent upcoming plans with lanes, bars, and markers.
• Quote
  • The Quote macro adds weight and credibility to your words with a visual cue, such as a vertical line on the left margin and a change in font color.
• Change history
  • The change history macro shows the history of updates made to a page. The information displayed here are version number, date and comment(author). It displays the information inline.
• Date
  • With the Date macro and the integrated date picker in an appealing calendar view, the current date or any other date can be quickly selected and integrated into pages.
• Anchor links
  • The anchor link macro allows you to hyperlink to a specific part of a page or live doc so that a reader can jump directly to that section upon selecting it. These can be especially useful for allowing your readers to navigate to specific parts of a long document. Anchor links are invisible to the reader when on a page in view-only mode. This macro is used in combination with the link macro. The anchor link is set to a particular name that will be input to the link macro section with the # prefix and the display text. The display text will be visible to the reader. When the visible text is clicked, the reader will be taken to the part of the page where the anchor link is set and is invisible.
• Decision
  • This macro gives you a nice and quick way to format and highlight any important decisions taken, especially during meetings. This macro adds some nice formatting and allows you to create decision reports. Starting your meetings with a review of past decisions and action items can make a huge difference.
• Excel/Word/ppt/pdf
  • Embed Excel, word, ppt or pdf with respective macros. To use each of these files upload the respective file using the image macro and select the file. Later on, use the respective file type macro as above to display it.
• Chart
  • The Chart macro allows you to display a chart based on tabular data. You can edit the macro parameters in the macro browser to configure the format of the chart.

Formatting toolbar & Content status

The formatting toolbar provides tools to format and color/highlight page content, create lists and tables, indent and align text, and insert other content into the page such as symbols, links, images, multimedia files, and macros. If a space admin allows statuses in a space, you can add or change them at the top of your content when editing. That means anyone with edit permissions can set the content status, choosing from the suggested options defined by the space admin or creating your own. Assign a status to your content to help set clear expectations for viewers and collaborators when they visit a page, live doc, or blog post.

Comments & Labels

For a page that is being prepared, you can use the comments feature to highlight the text and address the concerned team member with your comments. You can add comments to content while you’re editing it or viewing it in published status.

Labels are keywords that you can add to pages, live docs, and attachments to make them easier to group and find. For example, you could add the operations label to all pages related to operational activities. That would allow you to easily see, display, and search for related pages. It is handy for finding content across spaces if labels are consistently used. Labels are not only important to search results but can be extremely beneficial in dealing with certain macros. Since labels provide Confluence with another way of grouping information, you can use macros like Filter by label or the Labels list for more functionality and to easily navigate between content by looking for a topic or other articles if one does not meet your requirements. The Popular Labels macro can be used to generate the heat map of the most popular labels.

Whiteboard

Whether you’re brainstorming, planning projects, having a sprint retrospective or mapping out strategies, whiteboards provide a flexible canvas for collaboration. Use existing templates, sticky notes, shapes and connectors to visualize your thoughts and keep everyone on the same page. With real-time collaboration, your team can contribute simultaneously, making it easy to capture everyone’s input. Use a timer if you want to limit the sessions like in a retrospective.

If you have both a Confluence and Jira app in your Cloud site, you can seamlessly track project updates across both tools and create Jira work items without leaving the Confluence content you’re working on. To display Jira work items list on a Confluence page, use the Jira work items macro. Display a single work item, a list of work items, or show the total number of work items using separate macros. The simplest way to add a Jira work item to Confluence is to paste a Jira URL on a Confluence page or doc.

Further Reading:

• Jira Data Visualization: A Guide to Effective Dashboards
• Streamline Your Workflow with Jira Kanban Boards
• Jira Service Management Forms: Enhancing User Experience
• Getting started with Excel Macros
• Power Query Tips & Tricks

Service Request with Approvals

The decision to use SR with approvals issue type depends on the complexity and risk associated with the service request. If external validation is needed to ensure accuracy, compliance, or proper execution, approvals should be implemented. If the request is straightforward and can be handled by the assigned agent, then an approval step may be unnecessary. The default “Service request with approvals” issue type will appear as follows.

Workflow Improvement

Since we will be calculating the SLA for the tickets using this workflow, to ensure proper calculation we will add a transition from “In progress” status to the “waiting for customer” status and vice versa in the workflow. In general, the Project admin role will not be able to edit the workflow, for this you will need to be a Jira admin. The changes are implemented in the next steps as follows:

Creation of Service Request Form

Now that we are ready with our issue/work type, the next step would be to create a form using the above issue type with the updated workflow. Let us create a simple native form without attachment.

The sections can be seen as shown in the below “Simple form” picture. Summary and Description are the inputs from the user, while the other sections are hidden from the user. Labels section would be needed to display the ticket on a kanban board, while the Approvers section need to be input with email id of the approving authority. Multiple people can be specified but anyone from the specified list can approve the request.

Approvers get an email once the user submits the form. The template of the notification mail can be seen as shown in the below picture. In the standard version and above of Jira, this template can be edited to include the description of the Jira ticket as well (appropriate variable needs to be included in the Content section). If you are using an external form as an attachment and want to reflect the description there onto the contents of the mail, don’t forget to select the link to jira section in the attached form edit option.

If you want to include a template for the description of the ticket, using automation would be one idea, but can be used for the tickets created manually by the agents. However, if you want to use a template for the user, an external form can be used with a preset input that is linked to the description field. Regarding approvals, it can be done by using the notification email as shown below or on the ticket by clicking on the Approve button.

SLA Rule Setup

Now that we are ready with the updated workflow and a form that is using the updated workflow, the next step would be to use the SLA feature on the tickets. The SLA rule can be set by navigating to the Project settings > SLA section. The Goals and Conditions need to be set appropriately.

Our final form will appear as follows. The user can access this form to enter the necessary information.

Further Reading :

• Jira Service Management Forms: Enhancing User Experience
• Streamline Your Workflow with Jira Kanban Boards
• Integrating Service and Software Workflows on a Single Kanban Board
• Integrating Jira SM Alerts into Kanban Workflow
• Understanding Data Center and Service Desk Operations

Once you’ve created a Jira dashboard, you don’t have to do anything else. Every time you reload the dashboard, all the reports will update automatically with the latest data. You can also configure them to refresh at regular intervals. This makes dashboard reporting a much faster way of reporting. In this blog post, I will pick a few gadgets that is natively available in Jira. While these gadgets are really easy to use, they’re limited in what data they display, and in how they display it. Using an add-on is recommended if you want to make use of more features.

Creating a New Jira Dashboard

Navigate to Dashboards > View all dashboards and click on Create dashboard, or the option of directly navigating to Dashboards > Create dashboard is also there. This will open the following window. You can specify the Dashboard Name(TestDashboard), Description and set the Viewers, Editors permissions as well.

The next screen,i.e, the following one will appear once you click on the save button in the above window.

Once you are in this screen, you will have to plan the layout based on the target audience. An appropriate layout can be selected by clicking on the “Change layout” button at the top right. Let us use a pie chart to display the status of the sub-tasks, a table to display the volume of work handled by each assignee, a list that displays each of the work summary and a section that describes the contents of the complete dashboard.

Adding the Gadgets: Gadget 1

Now, let us start with the pie chart preparation. For this, you need to click on the “Add gadget” button at the top right and search for the pie chart gadget as shown below. Click on the Add button to add the gadget to the layout.

In the gadget, you need to add the project/filter information. For this, you need to first create a filter that can be used in the gadget. The filter “Sample Filter” is as shown in the figure below.

Once the filter, “Sample Filter” is ready, you can add it to the gadgets for display on dashboard. Now, our pie chart gadget, will look as follows once the filter “Sample Filter” is added.

Adding the Gadgets: Gadget 2

Additionally, one more gadget (Two Dimensional Filter Statistics) is added to display the volume of work handled by each assignee and can be seen in the below picture.

Adding the Gadgets: Gadget 3

One more gadget (Filter Results) is added to display the details of work handled by each assignee. Refer to the following picture for the settings to be done on the gadget. Once done, click on the Save button.

Adding the Gadgets: Gadget 4

Our final section will be a gadget that will be used to describe the contents of the complete dashboard. To do this, you first need to create a ticket (issue type bug is used) and add a description table, as shown below.

The next step would be to create a filter “test filter” to filter out the above bug issue type ticket. The JQL used to filter out the above ticket with label as “test” is as follows.

This filter will then be used in a separate gadget (Filter Results) to display the description contents of the above bug type ticket. This is done by making the below settings.

The Final Dashboard

Based on the settings done till now, our final dashboard will appear as follows. If you are sharing the dashboard with others in your team, see to it that the above filters used, have the necessary Viewers/Editors permissions set before sharing.

Further Reading :

• Data Visualization : MS Excel Dashboard Basics
• Data Visualization : Google Data Studio Dashboard Basics
• Microsoft Power BI : Getting Started Guide
• Getting Started with Power Query
• Navigating Jira : From Basics to Advanced

Every bit of energy we spend straining to recall things is energy not spent on the thinking that only humans can do. This includes inventing new things, crafting stories, recognizing patterns, following our intuition, collaborating with others, investigating new subjects, making plans, and testing theories. It’s time to acknowledge that we can’t “use our head” to store and recall everything we need to know. We need to start outsourcing the job of remembering to intelligent machines.

For modern, professional notetaking, a note is a “knowledge building block”. It is a discrete unit of information interpreted through your unique perspective and stored outside your head. Once you digitize your notes and observations, you can search, organize, sync them across all your devices, and back them up to the cloud for safekeeping.

As this way of working with information continues over days, weeks and months, the way your mind works begins to change. Your second brain becomes like a mirror, teaching you about yourself and reflecting back to you the ideas worth keeping and acting on. Thus, your mind starts to lean on it to remember more than you ever could, resulting in more productivity and also a fulfilling life.

Creativity

Creativity is about connecting ideas together, especially ideas that don’t seem to be connected. By keeping diverse kinds of material in one place, we facilitate this connectivity and increase the likelihood that we will notice an unusual association. These diverse kinds of material include, screenshots from an interesting Youtube video and a tweet sitting next to a quote from a philosophy book. An audio memo might be saved next to a link of a helpful website etc. Digital notes are virtual tools that you can use to turn vague concepts into tangible entities, which you can observe, rearrange, edit, and combine.

During brainstorming, the chances that the most creative and innovative approaches instantly come to our mind is not high. We tend to favor the ideas, solutions and influences that occurred to us most recently, regardless of whether they are the best ones. Now imagine if you were to draw on weeks, months, or even years of accumulated imagination in the form of second brain where lots of ideas can be permanently saved for long term. This support system can benefit us by coming out with the most creative and innovative ideas.

Three stages of progress

People who make use of the second brain ,i.e, the personal knowledge management system go through the following three stages as per the author.

• Remembering
• Connecting
• Creating

The first way that people make use of their second brain is as a memory aid. They use their digital notes to save facts and ideas that they would have trouble recalling other wise.

The second way that people use their second brain is to connect ideas together. Their second brain evolves from being primarily a memory tool to becoming a thinking tool. A piece of advice from a mentor comes in handy as they encounter a similar situation on a different team or situation.

The third and final way that people use their second brain is for creating new things. They realize that they have a lot of knowledge on a subject and decide to turn it into something concrete and shareable. Having a lot of support material gives them the courage to put their ideas out there and have a positive impact on others.

CODE

Capture, Organize, Distill, and Express (CODE) is a map for navigating the endless streams of information that we come across every day.

Capture: Keep what resonates

Every time we turn on our smartphone or PC, information immediately immerses us. However, we cannot consume every bit of the information stream. The solution is to keep only what resonates in a trusted place that you control, and to leave the rest aside. Thus, we can improve our ability to take better notes by training ourselves to notice when something resonates with us. You are what you consume, and that applies just as much to information as to nutrition. You can use capture tools to capture passages from ebook, excepts from online articles, quotes from podcasts to name a few. Refer to the link to find the best capture tools as provided by the author here.

Organize: Save for Actionability

Most people tend to organize information by subject. However, when it comes to digital notes, the best way to organize your notes is to organize for action according to the active projects that you are working right now. Organizing for action gives you a sense of tremendous clarity, because everything that you are keeping actually has a purpose.

Distill: Find the Essence

As per the author, once your notes are organized based on action, you will begin to notice patterns and connections between them. Every idea has an essence, i.e, the heart and soul of what it is trying to communicate. Your notes will be useless if you cannot decipher them in the future. By highlighting the takeaways, you will be able to remind yourself what the notes is about. Thus you are giving your future self the gift of knowledge that is easy to find and understand.

Express: Show your work

All the previous steps are geared towards one ultimate purpose ,i.e, sharing your ideas, your own story and your own knowledge with others. Most of us fall into the category of continuously force feeding ourselves with information. However, we never take the next step and apply this information. The author recommends us to shift our time and effort from consuming to creating.

PARA

The author refined, simplified, and tested an action-based approach to organize files and notes with thousands of people. This system called PARA can be used to organize four main categories of information in our lives: Projects, Areas, Resources, and Archives.

PARA can handle it all regardless of your profession or field. It organizes information based on how actionable it is, and not what kind of information it is. Organizing by actionability counteracts our tendency to constantly procrastinate and postpone our aspirations to some far-off future. PARA pulls these distant dreams into the here and now. It helps us realize that we already have much of the information we need to get started.

Progressive summarization

This is a simple process of taking the raw notes that you have captured and organized. You then distill the notes down to their most important points.

Layer one captures a few key excerpts of the article and includes a link to the original article at the bottom. To enhance the discoverability of this note, you can add a second layer of distillation. Here, the main points that provide hints of what the text is about needs to be bolded. For the notes that are especially long, interesting or valuable, the author suggests adding a third layer of highlighting. For this, you can use the “highlighting” feature available in most of the note-taking apps. This will often amount to just one or two sentences that encapsulate the message of the original source. A helpful rule of thumb that the author provides is that each layer of highlighting should include no more than 10 to 20 percent of the previous layer.

To avoid revisiting your notes and spending extra time deciphering their contents, this method can be highly effective. Focusing only on the bolded and highlighted points, as seen in layers two and three, makes it much easier to interpret the notes than rereading the entire article.

Intermediate packets

The idea of breaking down your work into smaller chunks is nothing new. However, it’s not enough to simply divide tasks into smaller pieces, you need a system for managing the pieces as well. The author calls the small pieces of WIP (Work In Progress) items as intermediate packets. The intermediate packets are the concrete, individual building blocks that make up your whole work. You can use any note as an intermediate step in a larger project or goal. Like LEGO blocks, the more pieces you have, the easier it is to build something interesting.

The author provides information on the five types of intermediate packets that you can use to create and reuse in your work.

• Distilled notes: Books or articles that you have read and distilled.
• Outtakes: The material or ideas that did not make it into your past project.
• WIP: The documents or other materials that you produced during your past projects.
• Final deliverables: Concrete pieces of work you have delivered as part of your past projects.
• Documents created by others: Knowledge assets created by your team or other stakeholders that you can refer to and incorporate into your work.

I guess our ability to quickly tap into these creative assets and combine them into something new will make all the difference in whatever we do.

Process of retrieval

There is no single, perfectly reliable retrieval system for the ideas contained in your notes. Instead, there are four methods suggested by the author that overlap and complement one another. The methods are as follows:

• Search: You can use the search function in your notes app as a quick, iterative approach to find the necessary information. If you’re unsure what to look for or interested in images or graphics, you can start browsing.
• Browsing: The notes apps offer a variety of features that allow you to easily browse through the hierarchy of folders. This browsing can be based on date created, image only view etc. Some of the notes may end up being useful in unexpected ways, which is where the tags will be helpful.
• Tags: Tags are the labels that you apply to certain notes regardless of where they are located. The main weakness of folders is that the ideas can get siloed from each other. Here, tags help in infusing your second brain with connections. Tags make it easier to see cross-disciplinary themes and patterns that defy simple categorization.
• Serendipity: This mysterious method is about creating the environment for ideas to jump out at you. The ideal conditions for the ideas to arise is to put all sorts of different kinds of material, on many subjects and in diverse formats, all jumbled together in your second brain. While using this method don’t restrict your search to that specific folder, instead look through related category projects or resources. Because visual patterns amplify serendipity, the author recommends saving images as well. Lastly, sharing our ideas with others for feedback also introduces an element of serendipity.

Divergence and Convergence

A creative endeavor begins with an act of divergence. You open the space of possibilities and consider as many options as possible. The purpose of divergence is to generate new ideas, but if all we do is divergence, then we never arrive anywhere. However, convergence forces us to eliminate options, make trade-offs and decide what is truly essential. It is about narrowing the range of possibilities so that you can make forward progress and end up with a final result.

The model of convergence and divergence is fundamental to all creative work. It’s a pattern we can observe across any creative field. For example, writers diverge by collecting raw material for the story they want to tell. This is followed by converge by making outlines, laying out plot points, and writing a first draft. We can overlay this model with the four steps of CODE to come up with a powerful template for the creative process. The first two steps of CODE ,i.e, Capture and Organize, make up divergence, while the final two steps, Distill and Express, are about convergence.

Divergence and convergence are not a linear path, but a loop. Once you complete one round of convergence, you can take what you have learned right back into a new cycle of divergence.

Routine Actions

The second brain is a practical system for enhancing your productivity and creativity. “Being organized” isn’t a personality trait that you are born with, nor is it merely about finding the right apps or tools. It is a habit of a repeated set of actions you take as you encounter, work with and put information to use. The author recommends three types of habits to incorporate into your routine. These habits help keep your second brain functional and relevant over time.

• Project checklists: Ensure you start and finish your projects in a consistent way, making use of past work.
• Weekly & Monthly reviews: Periodically review your work and life and decide if you want to change anything.
• Noticing habits: Notice small opportunities to edit, highlight or move notes to make them more discoverable for your future self.

When you turn your digital notes into a working environment rather than just a storage space, they become much more useful. As a result, you naturally start spending a lot more time there. Since this will turn out to align with your real needs, you need to use the habit of “organize as you go”.

Your second brain starts as a system to support you and your goals. You can easily leverage it to support others and their dreams. In doing so, you become a force for good in the world.

You can evaluate your current notetaking proficiency by making use of the free assessment tool provided by the author here.

Further Reading:

• Book Review : Getting Things Done
• Book Review : The Life-Changing Magic of Tidying Up
• Mind Management Basics
• Book Review : 7 Mindsets for Success, Happiness and Fulfillment
• Vision Board : A Tool To Manifest Your Dreams

In your service project, agents can raise different request types, such as “Get IT help” or “Suggest improvement,” among others. These are visible on the customer portal and are the means through which customers interact with the service desk. They are user-friendly and enable your agents to assist customers more effectively by automatically sorting requests into the correct IT service management work category (Service requests, Incidents, Problems, Changes, or Post-incident reviews). Each request type is based on an issue type, and a single issue type can be the basis for many different request types.

Features

When Jira admins or project admins add a form as part of a request type, it’ll appear as one form for customers to fill out when they raise a request. Doing this lets you include conditional sections, rich formatting, and custom layouts to forms. Some of the beneficial features of forms that you can make use of is as follows.

• Enhance your requests by attaching forms to request types. You can create a native form that appears on the portal with the limited options available on request types. To enhance your requests with various options, use forms and attach them to request types. You can either create a form from scratch or make use of the existing templates.
• The dynamic forms feature helps to create forms that use conditional logic to dynamically show or hide fields, and add headings, tables, and rich formatting.
• Add external resources, utilize the slash (‘/’) command feature. On the service desk portal home page, you can specify the documents or reference links under the title Related resources. In some cases, you would like to specify links directly on the form that you are working on. In such cases, you can use the slash command, commonly found in Confluence, in the forms to display a similar menu. You can select the link option from the dropdown to add links directly to your form.
• Allow users to edit their tickets after submission with the agents help. Editing is possible as long as the option “Lock this form once it’s submitted so only admins can edit” is disabled. Then agents can reopen the ticket and navigate to the following section in the ticket,i.e. , Attached forms > Three dots > Open customer edits option. This will enable the customer edits.
• Allow the users to edit the tickets anytime on their own. To allow this, the “Keep this form open for edits” option should be enabled in the form settings. After submitting the ticket, the form will continue to be in the Open status and the customer will be able to edit it at any moment. When the customer finishes updating the form, they must save it for the changes to be reflected on the created ticket.
• Use the “Save a PDF version of this form each time it’s resubmitted” to keep track of any changes to the resubmitted form. This is helpful to review the modification history of an issue over time.
• Collect more information by attaching forms to issues. Internal and external links. You can add forms to existing issues (Example: Service requests, Incidents, Changes etc. ) and choose to make them either internal or external(Visible to customer). You can use them to request more information from customers or to complete internal processes as the issue progresses.
• Create issues using a link to a form. You can create issues using a link that goes directly to a form without needing to navigate to the portal. This allows your team to quickly create requests on your customers behalf, share the link with others, or embed into a Confluence page for quick reference.

Create a request type

Let’s create a request type and then attach a form that was created separately. To start using forms, from your service project select Project settings, then select Request types. To view the Project settings, you need to be either a project administrator or a Jira administrator.

While creating a new request type you need to specify the Name, Portal group and issue type. The description section is optional. In our scenario, since it is an incident we will choose the issue type as incident.

Once you click on the Add button, following is the screen that appears. Here you can edit the sections of the form that is visible to the user or customer. You can drag and drop any of the suggested fields on the right side to the left side. Since we will be needing the attachment field we will be dragging it to the left. The form that we will create later can be added here, which will enhance the features of this native form.

Create a form

To start creating a form, navigate to the Forms section under project setting and click on the Create form button drop-down option. Select the appropriate option to create a new form from either the blank one or from an existing template.

If we are creating a new form from a blank one, click Create blank to see the below form appear. The options available are indicated in red. The top left rich text format section, the top right Add section button, the Add field button are all marked in red. In the content of the form, enter slash to display a list of options available that can further enhance the form features.

Now that we’re familiar with some basic features for getting started with forms, let’s create a simple form. The final form should appear as follows.

Since we have already created the request type, we will create a form that will enhance the request type sections (till Summary) to appear as above . First step would be to create a blank form by click on the Create form > Create blank option. Let us name the form as “Cybersecurity incident details”. The “Name of the user” section can be created using the Long text option under Add field. Since the red star mark is there, it will be a mandatory section to be entered. Refer to the below screenshot on where and how to set the fields.

Similarly for next text box ,i.e., the “PC model number”, all the steps will be the same, expect for the selection under Add field. Here you will be selecting Short text instead of Long text option.

The next section is the Maker name which is a drop down option. For this, under the Add field select the Dropdown field. Enter the Display name and Choices as displayed in the below screenshot.

The following section will be the start of the dynamic form section wherein we will use the “Radio buttons” option under Add field to decide on which section to display in the latter end of the form. If Yes option is selected one form will be displayed and if No is selected another form will be displayed.

In order to start using the dynamic form, first click on the Add section button. Fill out the sections as shown in the below picture ,i.e, Name, Show section, When field, Is any of. As you can see from the below picture that the Show section, When field and Is any of: fields are drop down sections. Select the appropriate option inorder to map to the earlier radio button query and answer ,i.e, either Yes or No. The below screenshot depicts the section to be displayed when “Yes” option is selected.

The text box below is the “Provide the details of incident”, which needs to be added from the Add field section by selecting the Paragraph option.

Now that only the “No” option form remains, click the Add section button again to create the form that appears when the “No” radio button is selected.

The sections below are the Date & Time , Paragraph fields available under Add field section. Name these sections appropriately so that they display correct names as above when the “No” option is selected. .

Once all sections are complete, click the Save changes button at the bottom to save the form.

Attach form to request type

To attach the form created earlier to the request type created much earlier, navigate to the Settings tab to the top left of the screen. Here we can link our form to the request type. Please refer to the image below for details on the option that needs to be enabled. One more way to attach a form to the request type would be to navigate to the request type that we created earlier and use the Attach form button.

The next step would be to access the form from the portal. For this, you need to navigate to the Channels > Portal option as shown in the screenshot below.

Further Reading :

• Microsoft Forms : Getting Started Guide
• Useful Microsoft Outlook Tips & Tricks
• AppSheet : The No-code App Development Platform
• Data Visualization : MS Excel Dashboard Basics
• Data Visualization : Google Data Studio Dashboard Basics

LLM

A language model is a probabilistic model of text. We can better explain this with an example.

I stay at Bangalore, and I contacted a travel agent to check for the nearest hilly region. The travel agent replied as ________________

Nandi hills, Kudremukh, Kumaraparvath, Mullayanagiri, Brahmagiri
Consider the word probability for each of the above as 0.3 ,0.2 , 0.1 , 0.1 , 0.2 respectively.

The language model knows about a set of words called vocabulary and it gives a probability to every word in its vocabulary of appearing in the blank. The LLM is no different from LM, the Large in the “large language model” refers to the number of parameters and there are no agreed-upon threshold upon which a model becomes large.

LLM Architecture

Here the focus is on two architecture models for LMs that are as follows:

• Encoders : Encoder models embed text, converting a sequence of words into an embedding. (vector/s or numerical representation). Example: BERT
• Decoders : Decoder models decode or generate text. Decoder models take a sequence of tokens and outputs next token. Decoder models tend to be pretty large compared to relatively small encoder models. Example: Llama2, GPT3, GPT4

LLM prompts

The simplest way to affect the distribution over the vocabulary is to change the prompt. A prompt is the text we provide to an LLM as input, which may include instructions and/or examples. Prompt engineering is the process of iteratively refining a prompt to elicit a particular style of response.

Some of the prominent prompting strategies are least to most, step back, chain of thought. However, we need to note certain issues associated with prompting, such as prompt injection/jailbreaking and memorization.

Another way to significantly affect the distribution over vocabulary is training. Prompting is simply changing the input to LLM. Since we keep the model parameters fixed, prompting alone limits the model’s distribution over words. Using prompting alone may be inappropriate when we have training data or need domain adaptation. In training, we are changing the parameters to the model. There are many ways to train a model or change parameters to a model. We can reuse the same cluster to fine-tune several models.

Training or LLM fine-tuning techniques

• Fine-tuning (FT) – Take a pre-trained model and a labeled dataset. Change all the parameters of the model for a task.
• Parameter efficient FT – Less expensive. Isolate a few parameters for training or add a set of new parameters to the model.
• Soft prompting – Add parameters to the prompt to force it to perform specific tasks during training.
• Continual Pre-training – Similar to FT but does not require labelled datasets. Continually training the model to predict the next word is done here. This is useful to adjust to new domain.

Decoding

This is nothing but taking the distributions and generating text. In other words, this is the process of generating text with an LLM. Decoding happens iteratively 1 word at a time. At each stage of decoding, we use the distribution over vocabulary and select one word to emit. The word is appended to the input and the decoding process continues.

Greedy decoding and non-deterministic decoding are the two types.
Greedy decoding is picking the word in the vocabulary with the highest probability. In non-deterministic decoding, randomly picking the high probability candidates is done. Accordingly, the next set of words vary based on the earlier choice and the probability score also changes.

For the below earlier example, greedy decoding will yield Nandi hills, whereas non-deterministic decoding may yield any random choice that can be even Kumaraparath.

I stay in Bangalore and I contacted a travel agent to check for the nearest hilly region. The travel agent replied as ________________

Nandi hills, Kudremukh, Kumaraparvath, Mullayanagiri, Brahmagiri
Word probability for each of the above are 0.3 ,0.2 , 0.1 , 0.1 , 0.2 respectively.

When using the non-deterministic decoding there is an important LLM parameter to know about, which is the temperature parameter. This parameter modulates the distribution over vocabulary and hence the output of LLM also varies accordingly.

Considering the above example, if temperature is increased the distribution is flattened over all the words and the probability of each words appears closer to each other. In case the temperature is decreased, the distribution is more peaked around the most likely word which is Nandi hills ,hence deviates more towards greedy decoding.

Hallucination

It refers to LLM generated text that is non-factual and/or ungrounded. There are some methods that are claimed to reduce hallucination (RAG). However, there is no known method to reliably keep LLM from hallucinating.

Grounding

It refers to generated text by LLM that is grounded in a document if the document supports the text generated. The research community has embraced attribution/grounding.

LLM Applications

A few applications of LLM include:

• RAG – This is primarily used in QA where the model has access to (retrieved) support documents for a query. User input would be a question. A question transformed into query is used to search a database for a corpus of documents. Retrieved documents and questions are provided as input to LLM and the model will generate a correct answer. The model is non-parametric,i.,e the same model can be used to answer any question about any corpus.

• Code Models – In code models, training is done on code, comments, and documentation instead of written language. This has been largely successful since 85% of people feel using models like co-pilot more productive. Code completion might be easier than text completion, given the fact that code is more structured, repetitive and less ambiguous than natural language. However, it has a narrow scope since writing code from scratch is fine but debugging success rate is less than 15 percent.

• Multi-modal and language agents – These models are trained on multiple modalities, i.e, texts, images and audio. In case of diffusion model, LLMs generate one word at a time to form a text. However, diffusion models generate images all at once rather than 1 pixel at a time. Image is a fixed size with fixed pixels but generating words is not fixed. Language agent models are more suitable for sequential decision-making scenarios. Examples: Playing chess, browsing the web in search of an item, operating a software autonomously.

OCI Generative AI Service

It is a fully managed service that provides a set of customizable LLMs available via a single API to build generative AI applications. Some of the high performing pretrained foundational models provided here are from cohere and meta. This service also provides flexible fine tuning, i.e, create custom models by fine tuning foundational models with your own data set. Lastly, it provides dedicated AI clusters that can host your fine tuning and inference workloads.

How does Gen AI service work

Generative AI is built to understand, generate and process human language at a massive scale. Use cases are text generation, summarization, data extraction, classification and conversation.

Some of the pre-trained foundational models available in OCI are

• Generation – Generate text instruction following models (cohere:Command & Command-light, meta:llama 2-70b-chat)
• Summarization – Summarize text with your instructed format , length and tone (cohere:command)
• Embedding – Convert text to vector embeddings semantic search. Search in many languages and across languages (cohere multilingual models)

Fine tuning

A key feature of OCI generative service is to fine tune the foundational models on a smaller domain specific dataset. This specifically improves model performance of specific tasks and improves model efficiency. OCI uses the T-few fine tuning ,i.e, the parameter efficient fine-tuning technique to enable fast and efficient customizations.

Dedicated AI clusters

These dedicated AI clusters are GPU-based compute resources that host the customer’s fine-tuning and inference workloads. Generative AI service establishes a dedicated AI cluster which includes a specific pool of GPUs and a dedicated RDMA cluster network for connecting the GPUs.

OCI console

Once you login to the OCI console and navigate to Analytics & AI > Generative AI, you will reach the OCI generative AI dashboard. Some of the important sections on the dashboard include the Playground, Dedicated AI clusters, Custom models and Endpoints. Clicking on the Playground will bring up the below screen.

The model drop down option will show all the pretrained foundational models ,i.e, the Generation, Summarization and Embedding models available. For example, when you choose an input(model) and an example, then press the generate button, the system generates a text output because the generation foundation model was selected.

If you are not satisfied with the output you can go ahead and change the parameters on the right side of the screen. One of the main parameters to adjust is the temperature that will make the output more creative with the increase in value. Also you can check the output selecting different models from the drop down. Finally, when you are satisfied with the output, click on view code that will display the code for the generated output in Java and python. You can copy this code and integrate it into your own application.

One main point to note w.r.t OCI generative AI service is that it is currently hosted only in the Chicago, Frankfurt and London regions that one need to subscribe to inorder to use the service.

Before starting with the different model types, let us understand about tokens. Language models understand tokens rather than characters. The number of tokens depends upon the complexity of text. Simple texts 1 token/word. Complex texts 2 or 3 tokens/word.
Example : Friend 1token, Friendship 2 tokens (Friend and ship)

Generation models

• Cohere – Command : Model parameters(52B), context window(4096 tokens – sum of i/p & o/p)
• Cohere – Command-light : A smaller and faster version of command model. Model parameters(6B), context window(4096 tokens)
• Llama-2-70b-chat Model parameters(70B), context window(4096 tokens)

The parameters in the above models are as follows:

• Max output tokens – Max number of tokens model generates per response. OCI 4000 tokens.
• Temperature – This controls how creative the model should be. In other words, temperature is a parameter that controls the randomness of LLM Output. A temperature value of zero makes it more deterministic, i.e, limits the model to use the word with the highest probability, whereas an increased value flattens the distribution over all words.
• Top p, Top k – Two additional ways to pick the output token besides temperature. Selects from top k items probability(top k) or top tokens based on sum of their probabilities with p value being the reference(top p).
• Stop sequence – It is a string that tells the model to stop generating more content that can control the model output. Ex: a period(.)
• Presence/Frequency penalty – Assigns a penalty when a token appears frequently. This helps to get rid of repetition in your outputs.
• Show likelihood – Determines how likely it would be for a token to follow the current generated token. Every time a new token is generated, a number between -15 to 0 is assigned to all tokens. Tokens with higher value are more likely to follow the current token.

Summarization models

• Command Model from cohere – Generates a succinct version of the original text that relays the most important information. This is the same model used for text generation, but the parameters are different here.

This model has separate set of parameters as follows:

• Temperature – Same as for the text generation model. The value can vary from 1 to 5.
• Length – This indicates the approx length of summary. Choose from short, medium, long.
• Format – Whether to display the summary in a free form paragraph or in bullet points.
• Extrativeness – How much to reuse the input in the summary? Paraphrase(low) to verbatism reuse(high).

Embedding models

Embeddings are numerical representations of a piece of text converted to number sequence. Embeddings make it easy for computers to understand the relationships between pieces of text.

Word embeddings that are numerically similar are also semantically similar (meaning) ,i.e, for example monkey and gorilla will be more numerically similar than monkey and a lion. Likewise, similar sentences are assigned to similar vectors, different sentences are assigned to different vectors.

• Cohere embed-english-v3.0 , embed-multilingual-v3.0 – As the name suggests, the models support english and multilingual languages. The model creates a 1024 dimensional vector for each embedding. Hence, every embedding, whether it is a sentence, phrase or a paragraph get converted into the 1024-dimensional vector. And the model takes a maximum of 512 tokens per embedding.
• Cohere embed-english-light-v3.0, embed-multilingual-light-v3.0 – This is a smaller and faster version of the above model. This model creates a 384-dimensional vector for each embedding and takes a maximum of 512 tokens per embedding.
• Cohere embed-english-v2.0 – Previous generation model that creates a 1024-dimensional vector for each embedding and maximum of 512 tokens per embedding.

Prompt Engineering

Prompt is the input or initial text provided to the model. Meanwhile, prompt engineering is the process of iteratively refining a prompt for the purpose of eliciting a particular style of response.

• Completion LLMs – They use a large dataset of internet text to predict the next word, rather than to safely perform the specific language task the user wants.
• Fine-tuned LLMs – We use Reinforcement Learning with Human Feedback (RLHF) on a fine-tuned model to better align its behavior with human preferences and instructions.

Prompting types

• In context learning – conditioning an LLM with instructions and or demonstrations of the task it is meant to complete.
• k-shot prompting – explicitly providing k examples of the intended task in the prompt. Few shot prompting is widely believed to improve results over 0-shot prompting.

Advanced prompting strategies

• Chain of thought – provide examples in a prompt to show responses that include a reasoning step.
• Zero thought chain of thought – apply chain of thought prompt without providing examples.

Training LLMs from scratch would be demanding in terms of cost, data and expertise that would not be a good idea. So the next best idea would be to customize. The following methods can be used to customize your model.

• Incontext learning/Few shot prompting – User provides demo in the prompt to perform certain tasks. Popular techniques include chain of thought prompting. Limitation is the model processing context length is 4096 or smaller.
• Fine tuning – Improve model performance on specific tasks. Better than prompt engineering. Improve model efficiency.
• Retrieval Augmented Generation – Language model is able to query enterprise knowledge bases to provide a grounded response. RAGs do not require fine-tuning.

In Machine learning, inference refers to the process of using a trained ML model to make predictions or decisions based on new input data. However, in case of LLMs, inference refers to the model receiving new text as input and generating output text based on what it has learnt during training and finetuning.

Dedicated AI cluster sizing and pricing

To create a dedicated AI cluster to fine tune a cohere.command model, you need two large cohere units. However, to host this fine-tuned model, you need a minimum of one large cohere unit. As a result you need a total of three large cohere units. Fine tuning a model requires more GPUs than hosting a model (therefore 2 units used above). We can reuse the same cluster to fine-tune several models.

A person X wants to fine tune a cohere command model and after fine tuning, host the custom models. The steps involved will be as follows:

• Fine tuning a cohere command model would require two large cohere units.
• Let us consider that the fine-tuning job takes 5 hours to complete.
• One more consideration is that the fine tuning is done every week with the availability of new data.
• Finally, the fine-tuned model will be hosted and would require one large cohere unit.

Calculations:

• Minimum commitment that the service enforces would be, hosting 744-unit hours/cluster, fine tuning 1 unit hour/fine tuning job.
• Unit hours for each fine tuning, 2 units * 5hrs /week = 10-unit hours.
• Fine tuning cost/month would be 10-unit hours/week * 4wks = 40 * large cohere dedicated unit per hour price.
• Hosting cost would be 744-unit hours * large cohere dedicated unit per hour price.
• Total cost = (40+744) *large cohere dedicated unit per hour price.

OCI Generative AI Security

Customer data and model isolation ensure that access to customer data is restricted within the customer’s tenancy, preventing one customer from seeing another’s data.

OCI security services

• Leverages OCI IAM for authentication and authorization.
• OCI Key management stores base model keys securely.
• We store the fine-tuned customer model weights in OCI Object Storage buckets, where they are encrypted by default, and the keys are managed by the Key Management service.

Building blocks for an LLM Application

One limitation of standard LLMs is that they rely on the knowledge they were trained on, which may become outdated.

Retrieval Augmented Generation (RAG)

RAG addresses this by retrieving up to date information from external sources, thus enhancing the accuracy and relevance of the information it provides.

RAG consists of the following three components.

• Retriever – This component is responsible for sourcing relevant information from a large corpus or database.
• Ranker – This component’s primary role is to evaluate and prioritize the information retrieved by the retrieval system.
• Generator – This is a language model whose job is to generate factually correct and accurate human like text based on the input it receives.

RAG can be implemented using either of the following two methods.

• Sequence model – It considers all the documents together to generate a single cohesive response.
• Token model – Here, the system constructs the response incrementally, with each part reflecting information from the retrieved documents specific to that section.

Now, let’s explore the basic RAG pipeline used in natural language processing. The RAG architecture combines a retrieval-based component with a generative model to enhance the generation of text.

• Ingestion – The system ingests documents and indexes embeddings in a database for quick retrieval.
• Retrieval – System uses the input query to search through indexed data to find relevant information.
• Generation – From the top k results, system generates response based on the information retrieved.

Practical Application of RAG

Inorder to explore practical application of RAG, we will look into a question-answer-based chatbot session. The process unfolds as follows:

• Initial prompt
• In addition to the initial prompt, the model considers the entire chat history.
• We then combine the prompt and chat history to form what we call an enhanced prompt.
• We pass the enhanced prompt through an embedding model, which transforms the text into a mathematical vector (captures semantic meaning).
• The model finds the most relevant information to a query by performing an embedding similarity search. Here, the vector of our enhanced prompt is compared against a database of other vectors.
• The system retrieves document/s associated with the matching vector ID.
• Augment the initial prompt with the retrieved documents.
• The augmented prompt is fed to an LLM. The LLM combines the context of the conversation with specific retrieved information to produce responses that are precise and reliable.

RAG triad is helpful to reduce the risk of hallucination.

• Context relevance – How well the RAG responses are aligned with the context of the conversation.
• Groundedness – Chatbot’s ability to provide responses that are accurate and based on reliable info.
• Answer relevance – Provide responses that are not only contextually appropriate but also specifically answer the user’s questions.

Vector databases

LLMs with RAG augment the process by using an external database, which is a vector database. Here we learn about how that vectors and vector databases improve the user experience.

Some of the most important features that makes the vectors databases so desirable with LLMs are as follows:

• Accuracy – stores data semantic relationship too that traditional database may not capture.
• Latency – Optimized for quick retrieval of high dimensional data.
• Scalability – Traditional database struggles with the sheer volume and high dimensional nature. Vector databases are adept at handling high volume and high dimensional data.

Vector databases are a core component of RAG-based LLMs. They use a type of database optimized for storing and querying vectors instead of a traditional rule-based.

• LLMs without RAG – It relies on internal knowledge learnt during pretraining on a large corpus of text. It may or may not use fine tuning.
• LLMs with RAG – LLMs with RAG use an external database which is Vector database. This extends the capabilities of LLMs to what they have been trained on.

Nearest Vectors and Advantages

Deep learning embedding models generate vectors that represent the semantic content of data, rather than the underlying words or pixels. These vectors efficiently compute similarity between items.

Unlike relational database, vector databases are optimized for multi-dimensional spaces, where the relationship between data points is not linear or tabular but is instead based on distances and similarities in a high-dimensional vector space.

As an example, let’s consider three groups of words based on similarities, i.e, fruits, animals and birds. If a user inputs peacock, it is closest to the bird group. To calculate the distance, we commonly use dot product and cosine distance to evaluate how similar or different two text embeddings are. When you provide a query or word, we will calculate the nearest vectors.

• KNN – You can use the K-Nearest Neighbors algorithm to perform a vector or semantic search and find the nearest vectors in embedding space to a query vector.
• ANN – Approx Nearest neighbor algorithms are designed to find near optimal neighbors much faster than exact KNN searches, but accuracy is lower than KNN.

Some of the most used vector databases are , Chroma, Oracle AI vector search, Weaviate, Pinecone, Faiss.

Lastly, the following list provides clarity on the advantages of Vector databases usage with LLMs.

• Addresses the hallucination (factually incorrect) problem inherent in LLMs.
• Augments prompt with enterprise specific content to produce better responses.
• Avoids exceeding LLM token limits by using most relevant content.

Keyword search

Keywords match with the terms people search for when looking for products, services, or general information (e.g., Google search). Simplest form of search is based on exact matches of the user provided keywords in the database or index. It evaluates documents based on the presence and frequency of the query term. Example : BM25 algo

Semantic search

Through this the LLMs can understand the context and meaning (rather than keywords) behind the user input leading to more accurate responses.

The two ways you can leverage semantic search are dense retrieval and reranking.

• Dense retrieval: This enables the retrieval system to understand, and match based on the contextual similarities between queries and documents.
• Reranking: This improves the quality and relevance of results generated during the initial retrieval process and reorders them.

Using the OCI generative AI service you can access pretrained models or create and host your own fine-tuned custom models based on your own data on dedicated clusters. Langchain is a Opensource framework for developing applications powered by language models. LangChain offers multitude of components to build LLM based applications, including an integration with OCI generative AI service, vector databases, document loaders, and many others.

Further Reading :

• IT Security Vulnerabilities and Exploitation
• Essential Cybersecurity Practices
• A Beginner’s Guide to Oracle Cloud Infrastructure
• An Introduction to OCI Cloud Operations
• Oracle Autonomous Database Cloud : Getting Started Guide

Microsoft tools can be very effective in task management provided the right tools are used for the right purpose. Here, i will be taking you through a few of the most used tools for task management.

Microsoft Outlook

Microsoft Outlook is mainly an email application that allows you to send and receive emails. In addition to emails, it combines calendar, contacts, and tasks in one place.

If there are a few email tasks and work items to be handled, then Outlook can be used for task management using the tasks option. Outlooks tasks directly integrate with Outlook email, allowing you to create tasks from emails. In case of Outlook 365 you can access the Tasks option from the sidebar under the More Apps as shown in the figure below.

Once the tasks option is selected, you will be at the next screen wherein you will see the list of all your tasks that is created by you or shared with you.

Microsoft Calendar

As the number of email tasks, work items increase, the Outlook Tasks App might not be the right tool since the visibility on the progress of tasks might be time consuming to ascertain. This is where the MS Calendar comes handy. Navigate to the Calendar option from the sidebar. In this screen, access the Home > Task option to create a new task.

If you are creating a task from an email, just drag and drop the email on to the calendar icon in the side bar. In the next screen you can set the start, end date and time for the email task.

Microsoft ToDo

In case of Outlook 365, the Microsoft ToDo is integrated with Outlook. Click on the ToDo icon in the sidebar to open the MS ToDo app as shown in the figure below.

Use the “My Day” feature to focus on tasks you want to complete today. ToDo is great for quick task entry and managing personal ToDos. You can create separate lists of tasks for different projects and share them with others for collaborative task management. Since this tool syncs with Outlook tasks, all the Outlook tasks will appear in MS ToDo and viceversa.

Microsoft Planner

At present, there is no integration of Planner with Outlook 365, but you can find this integrated with Microsoft Teams. From the sidebar on MS Teams, click on the Planner icon to open the app as shown in the figure below.

This tool is a combination of MS ToDo and Planner combined into one. At its core, Microsoft Planner is a task management application. You can easily create to-do lists, checklists and projects. The Kanban, charts and calendar views are intuitive for anyone to use and give a clear view of the tasks at hand. Although the features are fewer than other project management tools, they are effective for simple projects.

MS Planner is easy to use, so you won’t have much of a learning curve even if your team hasn’t used Kanban boards before. Kanban boards visually depict work at various stages of a process using cards to represent work items and columns to represent each stage of the process.

The above Microsoft tools if used efficiently can help transform Task Management , thus improving your productivity.

Further Reading :

• Navigating Jira : From Basics to Advanced
• Useful Tools for IT Professionals
• Few More Useful Tools for IT Professionals
• Useful Microsoft Outlook Tips & Tricks
• Useful Microsoft Teams Tips & Tricks

Jira helps teams plan, assign, track, report, and manage work. It brings teams together for everything from agile software development to customer support. The combination of customizability, agile support, integration capabilities with other tools, issue tracking features, reporting tools, scalability, community support, and deployment options on cloud & on-prem has contributed to the popularity of Jira as a project management tool.

The different types of Jira versions include Jira Software, Jira Service Management, and Jira Core, each designed for different project needs. Jira Software supports software development, Jira Service Management caters to service and support teams, and Jira Core functions as a general project management tool.

Company/Team managed project

In Jira Software and Jira Service Management, you have the option to create projects as either team-managed or company-managed. Check the bottom left menu in your project to determine whether you’re in a team-managed or company-managed project.

• Anyone on the team can set up and maintain team-managed projects, making them ideal for autonomous teams that want to control their own working processes and practices within a self-contained space. Another benefit is that they are easier to configure, allowing teams to start quickly.
• Teams prefer company-managed projects for collaborating across multiple projects in a standardized manner, such as sharing a workflow. Company-managed projects involve more complex project configuration compared to team-managed projects, requiring Jira admins to configure screens and schemes.

Agile Project management

Agile project management, primarily used in software development, emphasizes flexibility and collaboration. It incorporates customer feedback throughout the project life cycle. The methodology takes an iterative approach to development, breaking down work into small, manageable cycles called “sprints” to focus on continuous improvement in the development of a product or service.

Jira software is an agile project management tool that supports any agile methodology, be it scrum, kanban, or your own unique flavor. From agile boards, backlogs, roadmaps, reports, to integrations and add-ons you can plan, track, and manage all your agile software development projects from a single tool.

Starting & Closing a sprint on Jira

Sprints only apply to Scrum boards and not to Kanban boards. The scope of the project and the capacity of your team determine the length of a sprint, which can vary from a week to a month. In order to complete a sprint, you must be a Jira Administrator or a user with the manage sprints permission. All subtasks and stories must be closed in the current sprint unless they are being carried forward to the next sprint.

Jira issue types are the basic units of work that must be completed. Depending on the volume of work, you can break an individual issue down into even smaller issues to make the larger piece of work more manageable. Typically, the hierarchical approach in Jira Software begins with epics, followed by tasks, stories, and bugs, and then by subtasks.

After completing sprint planning, enter the stories and subtasks for the upcoming sprint into the Jira backlog, along with their estimates. The epics can be linked to the stories at this stage or once there is clarity on the epics at a later stage.

Enter the minimum sections in the story as shown in the red-framed areas in the picture below.

Enter the story points section marked in the above picture if story points are the basis of your estimation. However, inputting the original estimate in days/hours/mins in addition to the story points, will double the total estimation for the story, so should not be entered. Input the original estimate for a subtask, since hours and minutes need to be logged.

Entering the description area for a story and subtask that explains the mission and goals to be achieved is a good practice. In other words, there will be clarity on the definition of done for a particular story/subtask.

At the end of the sprint, when you have prepared the backlog for the next sprint and have clarity on which subtasks/stories need to be closed or carried forward, you can close the current sprint and start the new sprint.

Tracking work on Jira

Jira provides essential time tracking capabilities. Users can track their worked time and log hours on tasks with the help of the Jira time tracking fields. The comments section should be entered to provide clarity on the tasks performed. Even the work log section has provision to input the description of work done along with time entry.

Jira thinks of time and story points in a certain way. Here’s the math it uses:

• A story point is equivalent to 8 hours.
• 1 week (or 1w) is the same as 5 days.
• 1 day (or 1d) is the same as 8 hours.

Jira time tracking helps to provide a clear understanding of the project progress, estimates, and deadlines.

Service Management

Jira Service Management (Jira SM) is Atlassian’s service management solution for all teams. It helps in

• Empowering every team to deliver great service fast,
• Brings visibility to work, and
• Accelerates the flow of work between development, IT, and business teams.

Jira Service Management provides default ITSM practices (such as standard workflows) so you can customize your service project to meet your company’s rules, regulations and requirements. Your new service project comes with a service catalogue, ITIL workflows and basic reports. You can also set up a knowledge base in Jira Service Management by adding Confluence to your site. Confluence, a web-based corporate wiki and an Atlassian product, easily links with Jira SM and Jira Software.

Jira SM encompasses service management practices like service request, incident, problem, change, knowledge, asset and configuration management. It features 8 standard issue types: Change, IT Help, Incident, New Feature, Problem, Service Request, Service Request with Approval, and Support. For service teams, standard issues represent and track different requests made by your teams and customers. You can find the service requests screen in the following picture.

After creating an issue, you can request additional information from your customer or team members. Attaching another form to the issue allows for more effective collection of this information. You can add any number of forms to issues by selecting “Add form” from the quick-add toolbar.

Forms

Forms in Jira Service Management are customizable interfaces you can use to collect data from users. That way you can make sure that service teams have all the details they need to resolve requests. We can categorize forms in Jira Service Management into two main types: static forms and dynamic forms.

Static forms are the most basic type of forms used in Jira Service Management. They consist of a predetermined set of fields that collect specific information from users. These fields don’t change based on user input; they remain constant regardless of how the form is filled out.

Dynamic forms, on the other hand, are more advanced and interactive. They can change in real-time based on the inputs or selections made by the user. This adaptability is made possible through conditional logic. Using it, we can show or hide fields, change field options, or even alter the form’s flow based on the user’s responses.

Automation

The automation activity allows you to focus on the work that matters, removing the need to perform manual, repetitive tasks by allowing your teams to automate their tasks, processes, and workflows.

Automation rules on Jira consist of three parts: triggers that initiate the rule, conditions that refine it, and actions that execute tasks on your site. Triggers, conditions, and actions form the building blocks of automation. Combine these components to create rules that auto-close old issues or notify specific teams when a release ships. The below picture shows the automation screen and an example scenario.

In the above scenario, an automation ticket creates a new task every 30 days under the issue section.

Filters

A filter is a saved issue search. Jira users can create and manage their own filters and filter subscription.

Quick filters lets you further filter the issues on your Scrum or Kanban board. Your board comes with two pre-configured quick filters for “Only My issues” and “Recently Updated”. You can create more quick filters using Jira Query Language (JQL) for which you must be a Jira admin, project admin or board admin.

In the above picture, you can find one manually created quick filters “ToDo”. The ToDo filter can be used to filter out the stories and subtasks in ToDo status.

This setting will appear under Active sprints section under quick filter. Refer to the below picture for more clarity ,i.e, two of the default filters, Recently Updated and Only My issues under quick filters. The other two filters are manually created ones.

Advanced search

The advanced search allows you to build structured queries using the Jira Query Language (JQL) to search for issues. JQL is not a database query language, even though it uses SQL-like syntax. You can specify criteria that you can’t define in the quick or basic searches. JQL can help gain key project insights and help you find not just issues but also important information of projects.

Add filter as favorite

Navigate to the Filters>View all issues section and enter the JQL to filter out your issues. In order to add this as favorite, just click on save filter option next to the JQL entry section.

This opens a pop up wherein you need to enter the name and description of the filter along with viewer and editor permission. Subsequently, clicking the save button will save this filter under starred filters. This appears under the default filters on the left side of the screen as shown in the above picture.

Reports & Dashboards

Jira has a range of reports that enable you to visualize trends about your project, versions, epics, sprints, and issues. It categorizes reports into four types:

• Agile reports
• DevOps reports
• Issue analysis reports
• Forecast and management report

Teams frequently use reports like burndown charts, velocity charts, sprint reports, and control charts to measure the sprint performance.

On Jira, the dashboard enjoys great visibility and has become the primary place in the system to view reports of Jira data and connected systems. You can add and arrange gadgets on a Jira dashboard. Also, you can share dashboard pages with a user, group, or the entire organization. This makes dashboards the perfect tool to create and share report pages dedicated to specific projects and types of reporting.

Further Reading :

• ITSM and System Management Tools : A Basic Guide
• JP1/ITDM2 : A Basic Guide
• ITIL 4 Foundation : Getting Started Guide
• Data Visualization : MS Excel Dashboard Basics
• Microsoft Forms : Getting Started Guide