AI agents
AI that actually ships.
I build AI agents and features that hold up in production, not just in a demo, and I bring that approach to your business, your data and your customers.
What I do
AI strategy
Where AI will help, where it will not, and what it will cost to run, before anything is built.
Agents and assistants
Multi-agent systems that use tools, hand work to each other and know when to ask a person.
LLM integration and RAG
Models from Anthropic, OpenAI, Google or open source, connected to your documents and systems.
Evaluation, monitoring and safety
Tests for quality, cost tracking, guardrails and privacy, so you can trust what ships.
How I work
- Find the job Pick the work where AI saves real time or adds real value.
- Prototype Try it on your own data quickly, and measure the results.
- Harden Add evaluation, monitoring, guardrails and cost controls.
- Ship Put it in front of people, learn, and improve it.
A good fit if
- You have an AI idea and want to know if it will really work
- A prototype works, but production quality or cost is the problem
- You want agents that use your data without leaking it
Questions
Which AI models do you use?
I choose per job: models from Anthropic, OpenAI and Google, or open models when privacy or cost calls for it.
Is my data safe?
I design for privacy first: the least access needed, no training on your data, and clear rules for how long anything is kept.
Will it be expensive to run?
I estimate running costs before I build and track them after launch, so there are no surprises.