​Build autonomous AI workflows with agents, tools, memory, and feedback loops

​Why Loop Engineering?

​AI agents are easy to demo. They are much harder to make reliable.

​The interesting work starts when you stop thinking about a single prompt and start designing the loop around it: what the agent sees, what tools it can use, what it remembers, how it checks itself, when it asks for help, and how it improves over time.

​This is a small meetup for software developers exploring that layer.

​What to Expect

​It won’t be a traditional talk, and it won’t be a fully structured workshop. Think of it as a practical walkthrough followed by an open engineering jam.

​We’ll start with a hands-on walkthrough of how I use loop engineering to build autonomous AI workflows using agents, tools, memory, and feedback loops.

​We’ll look at real workflows, the patterns behind them, and the trade-offs that show up when you try to make these systems actually work.

​Bring your laptop if you want to follow along. We’ll build a few simple loop-driven workflows during the session.

​Open Engineering Jam

​In the second half, we’ll open it up.

​Anyone experimenting with agents, long-running workflows, evals, memory, tool use, or autonomous systems is welcome to share what they’re building, compare architectures, ask questions, and trade notes with others working through similar problems.

​Who Should Come?

​Come if you’re building with Claude, Codex, LangGraph, OpenAI, MCPs, custom agents, or your own messy stack.

​Come if you’ve moved past “chatbot” and are trying to make AI systems that can plan, act, observe, recover, and keep going.

​The Format

​Small group. Practical demos. Real workflows. No polished conference talks.

​Bring a laptop, a workflow you’re curious about, or just a few good questions.

​Built by Builders, Backed by the Best

​We’ve hosted engineers, financial experts, designers, founders, and researchers from some of the world’s most forward-thinking companies. Here’s a look at who’s been in the room.