I'll build a custom AI agent that can use tools, remember context, and complete multi-step tasks autonomously — production-grade, not a demo.

What's included:

Agent design and workflow architecture

Tool integrations (web search, APIs, databases, custom tools)

Persistent memory across sessions

Prompt engineering and guardrails

Failure mode handling (loops, hallucinations, tool failures)

FastAPI deployment

I've built and shipped agent infrastructure used in production which include evalflow, playagent, and remembr are all live on PyPI. You're getting someone who builds these systems, not someone learning from tutorials.