Services

I design and build full-stack web and mobile apps, shipped end to end — polished, modern front ends — built in JavaScript and React — wired to scalable APIs on the cloud. I’m a senior engineer with 25 years of experience, and I lean on modern AI tooling to design and ship fast, without cutting corners.

When your product needs it, I go deep on AI too — LLM tool calling, agents, and retrieval, built for verifiability rather than demos.

If you’ve got something you want built and shipped, let’s talk.

What I build Link to heading

Full-stack apps

  • Polished front ends — JavaScript, React, React Native, and TypeScript interfaces designed to look and feel great
  • Cloud APIs & back ends — Node.js and Python/FastAPI services deployed to AWS or Google Cloud Run, with databases, auth, and the production glue that holds it together
  • Web and mobile — iOS and Android via React Native, from MVP to launch

AI features (when you need them)

  • AI assistants & chatbots over your data — tool calling, retrieval, and an eval suite so you can trust the answers
  • LLM features inside your existing app
  • Natural-language data interfaces — ask in plain English, get answers from your real data (text-to-SQL)
  • Embeddings & semantic search over text and images

Packages Link to heading

Full-Stack App or MVP — I design and build a complete web or mobile app: a polished front end, a cloud API and database, deployed and ready for users.

AI Assistant or Feature — an AI assistant over your data, or an LLM feature inside your existing app: tool calling, retrieval, an eval suite to keep it honest, deployed to your cloud.

Feasibility Sprint — a short, fixed-scope engagement to de-risk an idea: I scope it, prototype the core, and hand you a build plan and a fixed quote — before you commit to a full build.

Ongoing & advisory — continued development, code review, or architecture and AI strategy, hourly or on a monthly retainer.

How I work Link to heading

Discovery first, then a fixed-scope build, then deploy and handoff — with documentation and tests (and eval suites for AI work) so you can trust the results after I’m gone. No black boxes.

Proof Link to heading

I built playcall, a natural-language NFL stats chatbot: it uses LLM tool calling to turn plain-English questions into SQL, runs a multi-turn agent loop across local (Ollama) and hosted (OpenAI, Anthropic) models, and holds itself to a 66-case eval suite — because my rule for AI is don’t trust the model, trust the verifier.

I’ve also shipped full-stack apps to production: a React Native app on Google Play, and web apps on Google Cloud Run — FastAPI services, custom domains, and managed databases. I write about this work on the blog.

Let’s talk Link to heading

Tell me what you’re trying to build, and I’ll tell you how I’d approach it — and whether I’m the right fit.