Recently I have been building Menagerie a tool for dispatching many Claude Code agents to complete tasks in parallel while maintaining visibility of progress, status and outputs. agents can show you what they are doing to the web app via playwright and ask for help when the need it.

I think the future of engineering will be based on creating robust agentic systems and amid all the doom and gloom that can exist in the industry these days I want to paint a positive picture of the exciting engineering problems that actually lay ahead of us.

Along with this I'll describe what it has taken me to deploy a system like this, e.g. sandboxing, cursor compatibility, terminals over the network and also how i've made it fun to use (for example you can give your agent a cute avatar, and if you so choose also a top hat)

Letting AI generate your UI is fast and fun right up until you need it to implement something specific. Throw in strict design systems, your company's existing component library and responsive breakpoints, and the whole process usually falls apart. The good news is that there is a solution. We'll explore what works and what doesn't with agentic coding.

Our coding agents are getting smarter, but the workspace we drop them into is probably not enabling them to use their full potential. The decisions, failed attempts, conventions, and product knowledge are still crucial context scattered across people’s heads, Confluence, old conversations, and forgotten pull‑request comments.

During this talk, we'll walk through the shift from the “good old” prompt engineering to agent experience, then I'll show a practical AX experiment at my current company that led to a better workspace for both agents and the people using them.

What if an AI agent could customize its own UI outside the chat and update it on the fly? As humans, some tasks that we work on are better done visually than in text format. That's why I started this experiment, where I give an agent a skill and teach it how to work in a highly visual workspace it can extend and update with custom UIs. I'll walk you through how we designed a skill, a CLI, and a front-end app that work in tandem to make it possible. How we spent weeks figuring out the best and most minimal design system an agent can use to generate "slop-free" designs. The project is completely open-sourced and available on https://moi.computer/