AI Engineer with 5 years of experience building production AI/ML systems for the energy industry. I architect agentic RAG systems, predictive ML pipelines, and compliance automation that run in enterprise environments.
class AIEngineer:
def __init__(self):
self.focus = [
"Agentic RAG & Multi-Agent Orchestration",
"Predictive Maintenance & Anomaly Detection",
"MLOps & Production ML Systems on Azure"
]
def deploy(self, model) -> Production:
return model.notebook_to_production()AI-powered knowledge management platform for oil & gas engineers. 7-stage agentic RAG pipeline with human-in-the-loop review, verified citations, and zero hallucinations. Built for corrosion engineers working with ASTM, API, and NACE specifications.
Next.js 16 React 19 TypeScript Claude Sonnet Supabase pgvector Voyage AI Vercel
Predictive Agent — LSTM · Scikit-Learn · Plotly · Docker → Code · Demo
Compliance Agent — PydanticAI · DSPy · Mistral · FastAPI → Code · Demo
Anomaly Agent — Isolation Forest · Gradio · Time-Series → Code · Demo
Vision Agent — Qwen2-VL · Transformers · Gradio → Code · Demo
+ LangGraph → Refactored FunctionMessage patterns, Enhanced fine-tuning docs
+ Pydantic → Core library contributions
+ AutoGen → Fixed Azure AI Client streaming stability
+ CrewAI → URL validation for Azure Gateways
+ Transformers → Documentation improvements
AI/ML Core: PyTorch · Scikit-Learn · LSTM · Isolation Forest · Time-Series Agents: LangGraph · AutoGen · CrewAI · PydanticAI RAG: pgvector · ChromaDB · Voyage AI · LlamaIndex MLOps: Model Monitoring · Drift Detection · A/B Testing · CI/CD
Infrastructure Cloud: Azure ML · GCP Vertex AI · AWS SageMaker Containers: Docker · Kubernetes (AKS/GKE) IaC: Terraform · GitHub Actions
Data & Pipelines Processing: Python · SQL · PySpark · PostgreSQL Serving: FastAPI · REST APIs · Streaming Pipelines Domain: SCADA/Sensor Data · Feature Engineering
M.S. Artificial Intelligence — University of Colorado Boulder · Expected 2027
5 years building production AI/ML systems
From adaptive learning engines to real-time blockchain fraud detection to industrial predictive maintenance. I take models from notebooks to production.