Vadivel Mohanakrishnan (Vadi)

Software Engineer / Engineering Leader

Thanks for stopping by. My journey started 25 years ago with a simple goal: building products that matter. From eCommerce, Telecom, EdTech to FinTech, that passion has never changed, only the tools have.

These days, I am exploring AI and Generative AI to see how they can make product development smarter and faster. Combine that with Cloud platforms like AWS and Azure and a strong DevOps mindset, and the possibilities are endless.

Continuous learning drives everything here. You will find ongoing POCs, MVPs, and experiments because for me the best way to understand something is to build it. Along the way, there have been a couple of patents [US8706726B2] [US20100161539A1] and six years as a Microsoft MVP in .NET and SQL Server, but those are just milestones in a much longer journey.

Let's connect. Whether it's sharing ideas, collaborating on tech, or exploring opportunities I am happy to have a conversation.

Side Hustles

Project Pulse

A privacy first, Advanced RAG system for accurate, multi-document Q&A over personal files. It features a pipeline with hybrid retrieval (Semantic + BM25) and cross-encoder re-ranking to ensure precision. It offers flexibility, allowing users to choose between the speed of Groq (Cloud) and guaranteed local processing via Ollama, putting data control squarely in the user's hands.

Python FastAPI LangChain Qdrant RAG

Demo Article Code

Feed Refiner

Feed Refiner is a unpublished Chrome extension built to refine the LinkedIn browsing experience by filtering out unwanted content. It uses the Groq AI model to analyze posts in real-time based on customizable categories, giving users control to either blur or hide irrelevant updates and maintain a professional, focused feed.

JavaScript Chrome Extension API AI/ML

Article Code

Wordris

Wordris is a creative puzzle game that merges the mechanics of Tetris with the word building challenge of Scrabble. It offers players a fun, fast paced experience where they form words from falling letter blocks. This project serves as a compelling demonstration of prompt engineering, as it was built start to finish using only natural language prompts via Lovable AI.

Lovable AI React TypeScript Tailwind CSS Vite

Live Article Code

NotesBuddy

NotesBuddy began as a COVID lockdown side project to learn Android development hands-on. The goal: build a simple, intuitive notes app for organising thoughts effortlessly. Within months, it was desiged, developed, and published on Google Play, teaching me the full Android lifecycle: design, development, publishing, secure key management (hard lesson!), and rapid prototyping under constraints. Though retired after losing signing keys, NotesBuddy remains a milestone and a reminder that that best way to learn is by building.

JAVA Android SDK SQLite