Santhosh Kumar
Data Engineer III at Walmart | Microsoft Certified Trainer | Build with Azure | Pythonista | 12X Hackathon wins
Bengaluru, India
Actions
Santhosh Kumar aka Sandy is a Data Engineer by profession, has an everlasting curiosity to try out new technologies, and shares his experience at community events.
Delivered five conference talks at Microsoft London office, Microsoft Bangalore office, Microsoft Chennai office, The Leela Palace, Chennai, and 10+ sessions for Azure Developer Community, won twelve hackathons participating solo (~$14,600+) from Microsoft, Atlassian, GitHub, Citi Bank, and other startups.
He is also a Microsoft Certified Trainer (MCT) since 2021, teaching about Azure AI capabilities, and has the following certifications under his belt DP-203, DP-100, PL-900, AI-900, DP-900, AZ-900, and Lead for Azure Developer Community - Chennai region.
Once a coder, Always a coder - Sandy
Coder By heart!!!
GitHub Profile - https://github.com/Santhoshkumard11
Area of Expertise
Topics
Leveraging Azure OpenAI for Developer Efficiency through Atlassian Forge App
Learn about how to use Azure OpenAI at it's best to increase developer productivity.
List of components used,
Azure OpenAI Service - Make Chat Completion API calls
Jira Rest API - Access Jira ticket descriptions and assignee
BitBucket Rest API - Access repo commit messages
Confluence Rest API - Access page details (team details and logging pages)
Forge App Storage - Store the usage logs of each user
node.js - Building the code for Forge backend
Building and deploying an enterprise grade multi-agent using Azure Semantic Kernal and Model Context
This talk will provide a complete walkthrough of building and deploying a multi-agent enterprise-grade application with Azure Semantic Kernal and Model Context Protocol plugins.
There will be a live demo of the multi-agent application build with Python to demonstrate how easy is it to create and deploy multi-agent systems.
Semantic Kernal and it's possibilities are endless, connecting to various sources from multiple cloud providers, choosing local models, use existing agents from other cloud providers (AWS Bedrock) and the best part is it's an open-source SDK.
Boost Your Productivity: How, When, and Why to Use Copilot Studio
Let's go to the basics for a second, and think of our problem as developers. Why are we looking for help? Why do we believe LLMs can help us?
Copilot Studio is a powerful utility that meets all your needs as a developer. Let's go over every aspect of it and try to understand what it can help with.
Leveraging Azure OpenAI for Developer Efficiency through Atlassian Forge App
Leveraging Azure OpenAI for Developer Efficiency through Atlassian Forge App
Play the good old Space Invaders game with Azure Cognitive Speech Service
Replicated the Legendary Arcade Space Invaders from 1978 which is a shoot 'em up arcade game developed by Tomohiro Nishikado.
The game can be played via any web browser which supports WebAssembly. In addition to using your keyboard to control the game, now you can use voice commands to do the same actions. This implementation is done with Azure Cognitive Speech service.
- I’ll talk about the detailed end-to-end implementation of the game deployed in GitHub pages.
- Live demo in how it works and some challenges with solutions
* Python - General Programming
* Pygame - Writing Game Logic
* Pygbag - Python WebAssembly for everyone ( packager + test server )
* asyncio - Tried some asyncio code to await between Speech Service and command executions
* GitHub Actions - Convert Python code to WebAssembly and deploy it in GitHub pages
* Azure Cognitive Services (Speech Service) - Speech to Text conversion of the game commands
Power Automate your way to make payments from inside Microsoft Teams
There is this notion of fear when you enter your credit card details into a website.
What if we can eliminate the fear by providing payments from within the Microsoft Teams application?
Teams now have roughly 270M+ active monthly users, so if you have a shop available on the Microsoft AppSource and you need a payment solution, it would mostly be a redirect to an external page. Why is that the case, but not anymore, with this embedded checkout.
- I'll present a demo of the complete working of the application and also explain the setbacks faced during this development with Power Automate, Adaptive Cards, and Power Apps (UI magic)
- This talk will be focused on how you can integrate any payment checkout page into Microsoft Teams application
Services used:
- Power Automate
- Power Apps
- SharePoint
- Azure Functions
- Adaptive Card
MLOps with GitHub Actions vs Deploying in Azure Machine Learning
I started out as an ML developer and was deploying ML models in Azure. I was using Azure App Services with Flask, which served well for the purpose but now I wanted to implement something that could be consumed instantly and at a low cost.
I didn’t find any good source to accomplish this, so as any developer would do, I started using the trial and error method, it took a while for me to figure things out and make it work.
I faced a lot of hurdles and challenges while doing this, will discuss them in detail, it's always good to know when to use what.
Services used - Azure Function, Azure Machine Learning, and GitHub Actions
Make payments from inside Microsoft Teams
There is this notion of fear when you enter your credit card details into a website.
What if we can eliminate the fear by providing payments from within the Microsoft Teams application.
Teams now has roughly 270M+ active monthly users, so if you have a shop available on the Microsoft AppSource and you need a payment solution, it would mostly be a redirect to an external page. Why is that the case, but not anymore, with this embedded checkout.
- This talk will be focused on how you can integrate any payment checkout page into Microsoft Teams application
- I'll present a demo of the complete working of the application and
- I'll also explain the setbacks faced during this development with Adaptive Cards and Power Apps (UI magic)
Leveraging Power Platform to build an Enterprise-grade Skin Cancer treatment application
Skin cancer is the most common form of cancer, globally accounting for at least 40% of cancer cases and the most common type is melanoma skin cancer, which occurs in at least 2–3 million people per year.
- This talk will focus on the insider aspects of how to solve it with the solution built completely on Power Platform from Microsoft Azure, as well as other Azure serverless offerings
- I'll be presenting a demo and a code walk-through on how this entire application functions (source code is open sourced)
- I'll deep dive a little into the image classifier, the abstract-level flow goes something like this, employees/nurses in the hospital can add a new patient and upload their skin sample image into the blob storage which triggers a callback and goes through the ML model for initial screening, the result is mailed to a dermatologist for closure. Once this is done, an appointment is scheduled automatically if cancer is detected and these are recorded in SharePoint List. The dermatologist can view the slots booked and the patient details.
Finally, I'll talk about the problems I faced in actually getting the image from Power Apps to the Function App.
Global Azure 2025 - Bengaluru Sessionize Event
M365 Saturday Bangalore 2024 Sessionize Event
Microsoft AI Day 2024 Sessionize Event
Power Platform Developer Power Summit Sessionize Event
Experts Live India 2024 Sessionize Event
Global Azure 2023 - Bengaluru Sessionize Event
Global AI Bootcamp - 2023 @Chennai Sessionize Event
Az Community Conference 2022 (In-Person Only Event) Sessionize Event
Microsoft 365 Saturday 2022 @Chennai Sessionize Event
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top