I build end-to-end machine learning pipelines and deploy them as real apps. Lately I'm deep into GenAI and Agentic AI — building intelligent assistants with LangGraph, LangChain, and LLMs that can reason, use tools, and remember context across conversations.
I enjoy turning messy data into models, and models into apps people can actually use.
📍 Based in India • 📫 mechmondip@gmail.com
Languages
GenAI & Agentic AI
Machine Learning & Deep Learning
Data & Visualization
Deployment & Tools
| Project | Description | Tech |
|---|---|---|
| 🤖 AI Chatbot | An AI assistant with persistent conversation memory, multi-tool reasoning, and a ChatGPT-style thread interface | LangGraph, LangChain, Groq (Llama 3.1) |
| 💬 Sentiment Analysis App | Production-ready sentiment classifier with a fine-tuned BERT model | PyTorch, BERT, Streamlit |
| 🏠 Melbourne Housing Price Predictor | End-to-end regression on 34K+ records. CatBoost achieved R² = 0.863 | CatBoost, XGBoost, LightGBM, Streamlit |
| 📞 Telecom Customer Churn Predictor | Binary classification on 7K+ customers. 80% accuracy, ROC-AUC 0.84 | Scikit-learn, EDA, Streamlit |
| 🐶 Dogs vs Cats Classifier | CNN with MobileNetV2 transfer learning. 74% → 97% validation accuracy | TensorFlow, Keras, MobileNetV2 |
- 🤖 Agentic AI systems — multi-agent workflows, tool-using LLMs, autonomous reasoning
- 🧠 GenAI applications — RAG pipelines, fine-tuning, prompt engineering
- 🔗 LangChain & LangGraph — building production-ready AI assistants
- ☁️ MLOps & deployment — taking models from notebook to production
- 📧 Email: mechmondip@gmail.com
- 💼 Open to: Data Scientist, AI/ML Engineer, and GenAI Engineer roles## Hi there 👋