I am a data science professional with hands-on experience building end-to-end machine learning and AI systems, from experimentation and modeling to deployment and integration into real-world workflows.
My work spans machine learning, deep learning, NLP, generative AI, and data engineering, with a focus on practical, production-oriented solutions.
- End-to-end machine learning pipelines
- LLM-based applications and retrieval systems
- Production APIs and deployment-ready systems
- Applied MLOps practices
- Data engineering and ETL workflows
A full-stack AI platform for transforming large volumes of YouTube comments into structured, actionable insights.
Repository: https://github.com/saurabhchavan7/CreatorInsight-AI-ML-Platform
An AI-powered desktop application that captures, understands, and acts on digital information using a Gemini 3 multi-agent architecture with proactive intelligence, semantic memory, and Google service integrations. Built for the Gemini Hackathon Google DeepMind 2026.
Repository: https://github.com/SapnaSChavan/gemini-hackathon-2026
An intelligent conversational system leveraging RAG architecture for YouTube content analysis and interaction.
Repository: https://github.com/saurabhchavan7/conversational-rag-platform-youtube
End-to-end CECL-compliant credit risk framework on 3.8M Fannie Mae mortgage loans ($694B portfolio). PD/LGD/EAD modeling, Federal Reserve 2025 stress testing, Monte Carlo simulation (10,000 scenarios), and real-time loan scoring dashboard.
Live Dashboard: https://cecl-credit-loss-modeling.streamlit.app
Repository: https://github.com/saurabhchavan7/cecl-credit-loss-modeling
| Project | Description | Repository |
|---|---|---|
| CreatorInsight AI/ML Platform | ML, NLP, and GenAI system for large-scale comment analysis | https://github.com/saurabhchavan7/CreatorInsight-AI-ML-Platform |
| Mnemos | AI-Powered Multi-Agent Second Brain with Proactive Intelligence | https://github.com/saurabhchavan7/gemini-hackathon-2026 |
| Conversational RAG Platform | RAG-based conversational system for YouTube content | https://github.com/saurabhchavan7/conversational-rag-platform-youtube |
| CECL Credit Risk Modeling Framework | CECL lifetime loss estimation, PD/LGD, stress testing, Monte Carlo, loan scoring | https://github.com/saurabhchavan7/cecl-credit-loss-modeling |
| AWS SageMaker Flight Fare Predictor | End-to-end ML project deployed using AWS SageMaker | https://github.com/saurabhchavan7/AWS-SageMaker-FlightFarePredictor |
| Term Deposit Subscription Prediction | ML model to predict customer subscription likelihood | https://github.com/saurabhchavan7/term-deposit-subscription-prediction |
| Repository | Focus | Link |
|---|---|---|
| LangChain Retrieval Systems | RAG pipelines, document ingestion, and retrieval workflows | https://github.com/saurabhchavan7/langchain-retrieval-systems |
| LangChain Core Concepts | Models, prompts, chains, runnables, and structured outputs | https://github.com/saurabhchavan7/langchain-core-concepts |
| LangGraph Agentic Workflows | Agentic AI workflow, Graph-based orchestration | https://github.com/saurabhchavan7/langgraph-agentic-workflows |
| Repository | Focus | Link |
|---|---|---|
| Deep Learning Projects | Practical deep learning implementations | https://github.com/saurabhchavan7/Deep-Learning-Projects |
| Deep Learning Algorithms | Core deep learning algorithm understanding | https://github.com/saurabhchavan7/Deep-Learning-Algorithms |
| ML Algorithms Implementation | Classical ML algorithms from scratch | https://github.com/saurabhchavan7/ML-Algorithms-Implementation |
| Scikit-Learn for Machine Learning | Pipelines, estimators, and transformers | https://github.com/saurabhchavan7/Scikit-Learn-for-Machine-Learning |
| Object Oriented Programming (OOP) | OOP principles implemented in Python | https://github.com/saurabhchavan7/Object-oriented-programming-OOP- |
Email: chavan.sau@northeastern.edu
LinkedIn: https://www.linkedin.com/in/contactsaurabhchavan/
Portfolio: https://chavansaurabh.com/
⭐ Feel free to star repositories you find interesting!