AI Engineer with 4 years shipping production AI and backend systems and 10 years of client-facing operations. I build production RAG, agent workflows, eval pipelines, and FastAPI services. Remote, US-based.
Based in the Inland Empire, California. 4 years building production AI systems -- FastAPI backends, retrieval pipelines, agent workflows, eval harnesses, and public demos. 10 years of client-facing operations before that, which still shapes how I scope work, communicate tradeoffs, and deliver under real constraints.
I hold 21 certifications from Google, IBM, Microsoft, DeepLearning.AI, Duke, Vanderbilt, and Anthropic (1,831 hours total), but the main proof on this site is shipped systems: DocExtract AI, Jorge Real Estate AI, EnterpriseHub, and mcp-server-toolkit on PyPI. Open source contributor to LiteLLM, FastAPI, and pgvector-python.
I build AI systems that have to work inside real workflows: RAG pipelines, agent workflows, FastAPI services, and evaluation gates. Strongest public proof includes a Claude-powered lead qualification platform that processed 500+ leads, DocExtract at 95.5% F1, and mcp-server-toolkit on PyPI.
My strongest layer is the application layer around LLMs: retrieval, tool use, orchestration, backend APIs, eval harnesses, caching, and operational debugging. Across public production repos, I have 3,500+ automated tests and a consistent bias toward reliability over demo-only polish.
The strongest non-code part of my profile is delivery ownership. I came from 10 years of client-facing operations, so I am comfortable translating vague requirements into working systems, communicating constraints clearly, and making AI useful in live environments.
Understand your data, existing systems, constraints, and what "done" looks like. No scope creep — explicit deliverables before code starts.
Design the system: data flow, component boundaries, API contracts, caching strategy, and deployment plan. You review before implementation.
Test-driven development. Tests first, then implementation, then refactor. CI runs on every push. You see progress in real-time via GitHub.
Deployed with Docker, documented with examples, demo mode included. Handoff includes architecture docs, test coverage report, and a walkthrough call.
Python, FastAPI, REST APIs, PostgreSQL, Redis, Docker, GitHub Actions CI/CD, GoHighLevel CRM integration
pytest, pytest-asyncio, RAGAS, LLM-as-judge, adversarial fixtures, GitHub Actions CI/CD, OpenTelemetry, structured logging
SQL, PostgreSQL, pgvector, Redis, embeddings, semantic search, BM25, reciprocal rank fusion, Streamlit, Plotly
Claude API, OpenAI API, Gemini API, RAG pipelines, agent workflows, tool use, structured output, LangGraph, prompt engineering
Open to remote AI Engineer, Applied AI, AI Backend, and LLM Platform roles. Selectively open to forward-deployed work that still centers on building and shipping real systems.