Data Scientist & ML Engineer building production ML systems at the intersection of energy infrastructure and modern AI.
I don't just train models — I deploy them. My work spans the full stack from raw data and feature engineering to cloud-hosted APIs, interactive frontends, and stakeholder-facing tools used in real capital planning decisions.
| Project | What it does | Stack |
|---|---|---|
| Cost Estimator | Parametric ML cost estimator for gas pipeline projects · AACE Class 5 compliant · live in production | Azure SWA · Azure Functions · scikit-learn · BLS PPI |
| CriticalPath AI | P6 Schedule Intelligence Assistant · ask critical path, float, slippage & budget questions in plain English | Azure OpenAI · NetworkX · Azure Functions |
| Receipt OCR | End-to-end receipt intelligence pipeline · extracts structured data · 94% accuracy · $0.006/receipt | Claude Vision API · FastAPI · Google Cloud Run · Sheets API |
| InformationExtraction | PDF RAG API · upload any document, ask questions, get grounded answers | FAISS · Sentence Transformers · DeepSeek · GCP Cloud Run |
| Cancer Prediction | PyTorch neural network · 97.4% test accuracy · breast cancer malignancy prediction | PyTorch · Streamlit · Plotly |
| Cold Email Generator | AI cold email writer · paste a job URL + upload resume → personalized email in seconds | LangChain · Llama 3.1 70B · Groq · Streamlit |
ML & AI → scikit-learn · PyTorch · LLMs · RAG · GraphRAG · Computer Vision · OCR
Cloud → Azure (Functions, Static Web Apps, Blob Storage, OpenAI) · GCP (Cloud Run, Vertex AI)
Languages → Python · JavaScript · SQL
Data → pandas · numpy · PySpark · Power BI · Power Query
MLOps → Docker · Azure CI/CD · GitHub Actions · FastAPI
- 6+ years turning messy operational data into decisions at a California's leading energy utility
- I build things end-to-end — model training, API deployment, and the UI that sits in front of it
- Background in Mechanical Engineering → MIS/Business Analytics → ML Engineering
- I believe the best ML work is the kind that actually gets used
Currently open to ML Engineer and Data Scientist roles. If you're building something that matters, let's talk.
