Hello, I'm
|
I build production-ready Generative AI applications and scalable software systems, working across LLM-powered agents, RAG pipelines, backend APIs, data pipelines, and modern web interfaces to solve real-world problems.
I'm a software engineer who loves turning "this seems impossible" into "wait, it actually works." Currently pursuing my Master's in CS at Northeastern, I've built automation systems at FedEx that handle 400+ stations (saving teams hours every day), launched AI agents at Lyzr AI that helped double demo conversions, and created full-stack apps that make a real-world impact.
I jump between building Spring Boot APIs, designing React interfaces, engineering RAG pipelines with LangChain, and writing technical docs and blog posts that thousands of people have read. Whether it's optimizing a slow database query, debugging a tricky race condition, or architecting a new feature from scratch, I love the challenge of making complex systems actually work.
What drives me? The thrill of learning new things and making them work in production. And honestly, just being curious about how things work and figuring out how to make them better.
Multimodal AI Agriculture Assistant
Built a FastAPI multimodal AI system with 6 specialized agents using LLM-based semantic routing, achieving 88% intent classification accuracy. Architected a RAG pipeline with FAISS for subsidy retrieval and integrated Llama 3.2 Vision for crop disease diagnosis.
Map-based Disaster Management Platform
Built a disaster management platform with Spring Boot backend and React frontend, reducing reporting latency by 40%. Architected an Apache Kafka event-driven pipeline for real-time alerts across 50+ users with Azure App Insights analytics.
Theatre Content Management System
Developed a theatre CMS using C# ASP.NET Core MVC and Entity Framework Core, reducing manual scheduling by 75%. Built with 5+ controllers managing 4+ entities and implemented 10+ role-based features.
Multi-Model Tag Prediction System
Built a multi-model tag prediction system using Hidden Markov Model, TF-IDF, and DistilBERT on 50,000+ Stack Overflow posts. Benchmarked traditional vs. deep learning architectures, outperforming baselines by 30% and reducing error rates by 80%.
Java MVC Application with Design Patterns
Architected a Java MVC system using Command and Strategy patterns for 15+ image operations including Haar Wavelet compression and Floyd-Steinberg dithering. Achieved 95% code coverage using SOLID principles and JUnit/Mockito testing.
Role-based Python Web Application
Created a Flask application with 3 role-based workflows, improving process efficiency by 35%. Designed a scalable MySQL database with CRUD principles and integrated triggers for notifications and alerts, reducing manual intervention by 20%.
Master of Science in Computer Science
Boston, MA
Bachelor in Information Technology
Chennai, India
"Living, Learning, and Leveling Up."