AI agents and workflow automation
Plan constrained agent workflows, connect tools, define approval boundaries, and make long-running work observable and recoverable.
Work With Me
AI engineering, backend systems, and product development.
Have an AI product idea, a backend challenge, or an existing system that needs improvement? I work with founders and engineering teams to design, build, and improve software systems.
Engineering problems I can help with
Retrieval, context, and AI memory
Shape ingestion, retrieval, evidence, and state so an AI feature can use relevant information without treating generated text as the source of truth.
Backend systems and integrations
Build and improve Python/FastAPI services, APIs, data models, asynchronous workflows, and external integrations.
Full-stack 0→1 product engineering
Turn a product problem into a working vertical slice across React/TypeScript interfaces, backend services, and deployment needs.
Architecture and reliability reviews
Review a proposed or existing system, make tradeoffs and failure modes visible, and produce a practical implementation plan.
Relevant examples are available in selected projects and the engineering blog.
Ways to collaborate
- Scoped engineering implementations
- Technical consulting
- Architecture and reliability reviews
- Contract engineering engagements
- Long-term technical collaborations
A collaboration usually starts with the goal, current constraints, and what is already in place. From there we can decide whether a focused review, a scoped build, or a longer engagement is the best fit. Scope, timing, and availability are agreed before work begins.
Questions
Can we talk while the idea is still taking shape?
Yes. A clear problem or user need is enough to start a conversation; a complete specification is not required.
Can you work with our existing engineering team?
Yes. Work can be scoped around implementation, architecture, or a focused technical problem alongside an existing team.
Can you improve an existing AI system?
That is a reasonable starting point. We can first understand its current behavior, constraints, and failure cases, then decide what work makes sense.
How do scope and pricing work?
We can discuss the problem and constraints first, then agree on a scope and engagement structure before implementation. I do not publish fixed prices or timelines because they depend on the work.
Do you take short or long engagements?
Possible formats include scoped implementations, consulting, architecture/reliability reviews, contract engineering, and longer collaborations. Fit and availability are discussed directly.
Have a technical challenge in mind?
A short description is enough to begin.