About
Hands-on experience across the full ML lifecycle including feature engineering, model validation, deployment, and Retrieval Augmented Generation. Strong Python and FastAPI background with emphasis on reproducibility, low-latency inference, safety-first thresholds, and real-world system reliability.
Work Experience
Skills
Check out my latest work
I've worked on a variety of projects, from low-latency ML services to enterprise-grade GenAI systems. Here are a few of my favorites.
Built an enterprise-grade GenAI assistant for AWS IAM using a robust Retrieval-Augmented Generation (RAG) pipeline over official documentation. Implemented hybrid retrieval (semantic + BM25) with BGE embeddings and Qdrant to ensure high-precision, context-grounded responses. Designed a scalable FastAPI backend with background indexing and a Next.js frontend, enabling low-latency, security-focused Q&A strictly constrained to AWS IAM knowledge.
Built an aircraft engine failure prediction system using XGBoost with 100+ time-series features on NASA CMAPSS data. Achieved ROC-AUC 0.996 and 75% recall with sub-15ms CPU inference using a FastAPI service.
Building to Win
I believe in constant growth and accountability. I participate in hackathons to pressure-test my skills under tight deadlines and strive to build winning solutions. It's about more than just building; it's about competing at the highest level.
Code to Unlock
VIT
3rd Place in International Techno-Management Festival.
Yantra Central Hackathon
VIT
Semifinalist among 150+ teams.