CS student at VIT Bhopal building things at the intersection of software engineering and AI — multi-agent RAG pipelines, full-stack platforms, and ML systems that solve real problems.

• 🎓 B.Tech Computer Science & Engineering, VIT Bhopal (2023 – 2027)
• 🤖 Deep interest in Generative AI, Agentic AI, RAG systems, and full-stack development
• 🧠 Strengthening DSA and CS fundamentals alongside AI/GenAI engineering
• 📬 Open to Software Engineering and AI Engineering opportunities

AI Engineer Intern  ·  Coding Blocks School of Technology  ·  June 2026 – July 2026

• Built an LLM-based Legal Research Assistant using LangChain, LangGraph, Qdrant, and Llama 3.3 for Indian Supreme Court judgment retrieval.
• Designed a multi-agent RAG workflow enabling semantic search, case summarization, and legal memo generation.

MedNexus — AI-Powered Clinical Intelligence Platform

Full-stack clinician dashboard with real-time chat, voice input, file upload, and async REST API communication.

• FastAPI backend + MongoDB for patient metadata + ChromaDB for semantic retrieval
• Automated ingestion pipeline for 4 document formats (PDF, CSV, TXT, scanned images via OCR)
• Modular LangGraph backend with 5 service modules: patient lookup, retrieval, ingestion, trend analysis, alerts

FastAPI React.js LangGraph LangChain MongoDB ChromaDB Python EasyOCR

Legal Research Assistant — Multi-Agent RAG Pipeline

Autonomous retrieval and analysis of Indian Supreme Court judgments using a 4-agent LangGraph pipeline.

• Planner → Retriever → Analyzer → Memo Writer with conditional replanning loop for retrieval quality control
• Indexed ~5,000 Supreme Court PDFs (2010–2024) into Qdrant using BAAI/bge-small-en-v1.5 embeddings
• Llama 3.3 70B via Groq with structured prompt engineering for JSON extraction from legal documents

LangGraph LangChain Qdrant Groq API Llama 3.3 Google Colab

SafetyNet — Crowdsourced Disaster Preparedness Platform

Full-stack disaster management platform for real-time incident reporting at scale.

• React + Node.js + Express.js supporting 500+ concurrent users
• YOLOv8 model for human detection under debris — 94% accuracy; ML models for flood & earthquake risk
• Geospatial REST APIs with PostgreSQL + PostGIS; spatial queries under 300ms; OpenStreetMap integration

React.js Node.js Express.js YOLOv8 PostgreSQL PostGIS Python TypeScript

Languages C++ Python JavaScript TypeScript SQL Bash

AI / GenAI LangChain LangGraph RAG Agentic AI Prompt Engineering HuggingFace Transformers Groq API TensorFlow Keras Scikit-learn OpenCV NumPy Pandas Matplotlib

Web / Backend React.js Node.js Express.js FastAPI REST APIs HTML CSS

Databases MongoDB PostgreSQL MySQL ChromaDB Qdrant

Developer Tools Git GitHub VS Code Postman Google Colab

• Flipkart GRiD 8.0 (2026) — Selected for Round 3, advancing through Flipkart's flagship engineering competition.
• LeetCode — 500+ problems solved, building strong DSA foundations for technical interviews.

Open to Software Engineering and AI Engineering opportunities 🚀