🚀 Data Scientist | AI Researcher | Generative AI Developer

I specialize in Machine Learning, Generative AI, LLM Fine-Tuning, and Model Deployment. With over 6+ years of experience, I've worked on cutting-edge AI applications, including static code analysis improvement using LLM, drug toxicity classification using multimodal approach and Finetuning deep learning models (LLM, VLM and pretrained) for custom usecases.

💡 I’m passionate about LLM fine-tuning, reinforcement learning, and AI-powered automation. My work involves building scalable AI applications, optimizing deep learning models, and deploying AI systems efficiently using FastAPI and Docker.

• LLM Fine-Tuning & RAG – Experimenting with LLM finetuning, LoRA, qLoRA, and multi-GPU training
• Reinforcement Learning & AI Agents – Implementing from scratch RLHF, PPO, DPO, ORPO, GRPO, OpenAI Agents and Google ADK
• Investment Portfolio Optimization – Designing a long-term strategy for wealth creation

Languages: Python (ML/DL), SQL
AI/ML Frameworks: PyTorch, Huggingface transformers, openai, TRL Library, XGBoost
Backend: FastAPI, Celery, Docker, Redis
Databases: PostgreSQL, MongoDB
Deployment: Git, GitHub Actions, Ubuntu, systemd, Nginx

✅ Implementing and learning from research papers related to AI
✅ Fine-tuning & training LLMs from scratch in a scalable way
✅ Reinforcement Learning (RLHF, PPO, DPO, ORPO, GRPO)
✅ Multi AI Agents and their applications
✅ Google Cloud Platform and backend development

🔗 LinkedIn
💻 GitHub
📧 Email: beheradinabandhu50@gmail.com

💡 “AI is the new electricity.” - Andrew Ng
🚀 Let’s build something awesome together!