Looking for a few serious testers for an AI agent workflow tool
I’m currently building Grace, a workspace for coordinating multiple AI agents and tools in structured workflows.
The idea is not just to chat with one model, but to create pipelines such as:
one agent plans the task
another implements it
another reviews the result
issues are sent back automatically
the process continues until predefined conditions are met
Grace is still under active development, and I’m specifically looking for a small number of testers who are willing to use it properly and provide detailed, honest feedback.
I’m especially interested in feedback on:
whether the workflow is actually useful in real projects
what feels confusing or unnecessary
where the automation breaks down
which integrations or controls are missing
whether the agent handoffs and review loops work as expected
This is not a finished-product launch, and I’m not looking for people who only want to click around for five minutes.
I’m looking for developers, technical founders, automation enthusiasts, or people already working with tools like Claude Code, Codex, Cursor, OpenClaw, or local LLMs.
If that sounds interesting, send me a DM with a short note about how you currently use AI tools and what kind of workflow you would test.
Sorry for formatting, written on mobile