Senior Data Scientist at Autodesk with 6+ years of experience building production ML systems that deliver measurable business impact. I specialize in the full ML lifecycle — from feature engineering and model development to deployment, monitoring, and stakeholder dashboards.
- 🏢 Autodesk → ML for cloud optimization, anomaly detection, LLM-powered tooling
- 📊 Expedia → Large-scale propensity modeling for 200M+ customers
- 🤖 Tata Elxsi → CNN-based health monitoring & deep learning R&D
- 📍 Bay Area, CA | 🎓 MS Computer Science, Penn State
💰 ~$4M annual cloud compute savings via cost-aware ML model at Autodesk
📉 12% reduction in customer opt-out rates at Affine Analytics
👥 200M+ customers served through personalized recommendation systems
🚨 Real-time API anomaly detection preventing production incidents
🤖 Text-to-GraphQL interface via LLMs for streamlined developer experience
Languages
ML / AI
Data & Cloud
MLOps & Tooling
Chat with Documents — Advanced RAG
A local RAG web app with two production-grade pipelines built on top of a swappable LLM/embedding/vectorstore backend.
Python LangChain RAG Vector Store Agentic AI Ollama OpenAI Anthropic
Snowflake MCP Server — Pure Async
A production-ready MCP (Model Context Protocol) server for Snowflake using the low-level async API — giving full control over server lifecycle, tool registration, and async execution.
| Feature | Details |
|---|---|
| 🔌 Tools exposed | execute_query, list_databases, list_schemas, list_tables, describe_table, check_database_exists |
| 🛡️ Query safety | Read-only validation (SELECT, WITH, SHOW, DESCRIBE, EXPLAIN only) |
| ⚙️ Production features | Persistent connection · health checks · timeout control · query tagging · cache control · row limiting |
| 📐 Config | Pydantic models for type-safe, validated configuration with clear startup errors |
Python MCP Snowflake Async LLM Tooling Pydantic
| Project | Description | Stack |
|---|---|---|
| 🫁 COVID-19 Chest X-ray Detection | Transfer learning with VGG16 for binary medical image classification | PyTorch · VGG16 · OpenCV |
current_focus = {
"LLMs": ["Fine-tuning", "RAG systems", "Agentic workflows"],
"MLOps": ["Feature stores", "Model monitoring", "SageMaker pipelines"],
"GenAI": ["LangChain", "LangGraph", "Prompt engineering"],
}Building ML systems that don't just work in notebooks — they work in production.