AI Full-Stack Engineer · B.Tech (IT) @ IIIT Allahabad '26
Applied AI systems engineering — from model design through agent orchestration to user-facing apps.

B.Tech (IT) at IIIT Allahabad ('26), working at the intersection of applied AI and shipped product. My focus spans model design and evaluation, agent orchestration, and the full-stack systems that surface them to users.

Recently completed a Technology Internship at ICICI Bank working on Applied ML for Observability & AIOps — built a banking-grade AIOps platform with ML-based anomaly detection; received a Pre-Placement Offer for full-time return. Current work centres on ongoing AI/ML research alongside an open-source agentic-RAG system recently shipped to PyPI.

Open to AI Full-Stack and ML Engineer roles where I can own the full vertical from model layer through API to interface.

• 🧮 financebench-rag-agent — Multi-agent RAG for financial document Q&A; 72.7% on FinanceBench (κ = 0.932 calibrated judge). Shipped to PyPI + GHCR. LangGraph · Qdrant · Claude · FastAPI

• 🔋 hybrid-ml-mcdm-battery-eol — ML + multi-criteria framework for EV battery end-of-life routing across EU / GBA / India regulatory regimes. SoH RMSE 2.43pp on 1,581 cells, 7 chemistries. Live Streamlit demo · manuscript in preparation. XGBoost · TCN · TOPSIS · BWM

• 🩸 edge-cbc-analyzer — 5-tier Complete Blood Count pipeline on Jetson Orin Nano. mAP50 = 0.988 at 2.7–3.3× fewer params than published SOTA. 8th-semester research project. YOLOv8-seg · EfficientNet-B0 · TensorRT FP16

• 🧘 zen-daily-wisdom — Personalized RAG with Thompson-sampling bandit ranking; recall@5 = 0.767, regret 47.1 vs 124.8 baseline. Deployed end-to-end. Next.js · FastAPI · pgvector · Gemini

• 🏦 banking-observability-platform — Open-source AIOps stack for 37+ banking microservices with ML-based DDoS detection. Built during ICICI Bank internship. Python · Docker · TensorFlow · Prometheus

• 🛒 amazon-greencart-hackathon ⭐ 58 — Sustainable e-commerce with XGBoost EarthScore, DBSCAN group-buying, multi-agent shopping assistant. Semi-Finalist · Amazon HackOn S5. React · FastAPI · LangGraph

• 📚 Ongoing AI/ML research — applied deep learning and edge inference, building on the CBC research pipeline.

AI / ML

Backend

Frontend

Data & Infra

• 🎓 CGPA 8.89 / 10 — B.Tech (IT), IIIT Allahabad
• 🎯 99th percentile in JEE Mains (12L+ candidates)
• 💻 600+ problems solved on GeeksforGeeks
• 🏦 Built ML systems in a banking environment during ICICI Bank internship (2025)
• 🥇 Multiple Junior Science / Cyber Olympiad medals and academic awards

Open to engineering opportunities. Reach out to discuss AI systems, agentic architectures, or applied ML.