Skip to content
View princ3kr's full-sized avatar

Block or report princ3kr

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
princ3kr/README.md

Hey, I'm Prince 👋

3rd-yr Electrical and ELectronics Engineering Student @ VNIT Nagpur | Gen AI Engineering · RAG · Knowledge Graphs · Multi-Agent Systems | ML Team Lead @ ACM VNIT

🔨 What I'm Building

R2G Mapper — Parses code repositories into graph structures using NetworkX + Neo4j + ChromaDB. Implements hybrid retrieval (vector + BM25 via RRF), cross-encoder reranking, and RAGAS-based evaluation. Multi-tenant via repo_id isolation.

GraphMASAL — Multi-agent adaptive learning system built on LangGraph + Neo4j. Two-stage retrieval pipeline, MSMS planner, TinyDB student state management. Think Notebook LM, but actually agentic.

VQA Model — ResNet50 + BERT on VQA v2.0. 58.56% soft accuracy. Multimodal, end-to-end.

🧠 What I Work With

Python PyTorch Neo4j LangChain HuggingFace FastAPI

RAG · Knowledge Graphs · Multi-Agent Systems · Cross-Encoder Reranking · Graph-based Retrieval

📌 Currently

  • Building R2G Mapper — turning codebases into queryable knowledge graphs
  • ML Team Lead @ ACM Student Chapter, VNIT — mentoring juniors in ML fundamentals

📬 Reach Me

I build things to understand them. Open to internships, research collabs, and interesting problems.

Pinned Loading

  1. Notebook-LM-Mini Notebook-LM-Mini Public

    Graph-based multi-agent adaptive learning system that transforms engineering syllabi into Neo4j knowledge graphs and generates personalized learning paths using MSMS pathfinding and LLM-powered dia…

    Jupyter Notebook 1

  2. Repo-Mapper Repo-Mapper Public

    Jupyter Notebook

  3. VQAModel VQAModel Public

    🖼️ Visual Question Answering model using ResNet50 + BERT with Gated Fusion. Achieves 58.56% accuracy on VQA v2.0 validation set.

    Jupyter Notebook 1

  4. City-Sync City-Sync Public

    AI-powered civic issue reporting platform with geospatial deduplication, differential privacy, and automated resolution verification for smart cities.

    Python 1