Coffee ☕ • Code ☢️ • Chaos ↱ • Repeat |
To Err is human, to git commit is divine
Welcome to my GitHub profile!
- 💼 Research Engineer @ IBM
- 🎓 CSE 2026 | IIIT Jabalpur
- 🤖 AI Research Intern @ IBM Research (May 2025 – Aug 2025)
- Worked on RLHF (Reinforcement Learning from Human Feedback)
- Hands-on experience with GRPO, PPO, and DPO training algorithms
- 📊 Data Science Intern @ Meesho (Jan 2026 – Jun 2026)
- 🏆 MLH HackCBS 7.0 (2024) Winner
- 🌸 MLH HackByte 2.0 (2024) Best All Girls Team
- ⚡ Winner at ThermaOracle, Cognizance, IIT Roorkee (24-hr ML Hackathon)
- 🎯 Among top 15 in Meesho DICE Challenge in Data Science Track
- ❤️ Passionate about AI, ML, Deep Learning, Data Science, and Model Alignment
- 🌱 Always exploring new technologies and solving challenging problems
- 🔗 Looking forward to connecting with people and learning more!
| Category | Technologies |
|---|---|
| Languages | |
| ML / AI | 🧠 Basics of ML/AI & Core Models • ⚙️ Systems in AI • 🧬 RLHF (GRPO, PPO, DPO) |
| Data Science | |
| Web Development | |
| Tools & Platforms |
- ABC Rule Miner: A supervised explainable AI rule-mining algorithm for extracting high-confidence IF–THEN rules for behavior prediction (Rule extraction, confidence filtering, coverage evaluation, and comparison with Apriori).
- RagMantra: A Retrieval-Augmented Generation (RAG) interface using local LLMs via Ollama for PDF summarization, document Q&A, interactive chats, and multi-document retrieval.
- Object Detection: A Streamlit-based object detection app built using Faster R-CNN.
- CourseCrafter: A web platform that helps users discover tailored courses matching their specific learning interests.
- 🔬 Advanced Deep Learning
- 💬 Natural Language Processing
- ⚖️ Model Alignment & RLHF
- 🌐 Full-stack Web Development
- 📐 Data Structures & Algorithms
I'm always open to networking, collaborations, or tech discussions.
- 💼 LinkedIn: Joyee Biswas
- 🐙 GitHub: @Joyee2004
I believe the best code is not just functional—but also elegant, readable, and thoughtful.


