Need Feedback 🙏 : We built A "Miro" canvas to co-work with your agents
Hey PHs 👋
Sharing something we've been building for the last few months to get feedbacks.
What it is: Denker.ai is a visual canvas where you can delegate tasks to AI agents — like assigning work to human teammates.
The problem we tried to solve:
Prompt chaining feels like babysitting, not delegating
Setting up a team of agents requires too much effort and config
Checking deliverables from agents, means opening more tabs and windows
What we've built:
A team of agents working out of the box, supported by our existing subscription
Visual canvas to map out what needs doing and deliver whats done
Agents pick up tasks autonomously (no step-by-step prompting)
They ask clarifying questions when stuck, just like a real teammate would
Where I am:
Live with some testers
No traction yet, just getting started
Planning to launch officially in 6 weeks
What I'd love feedback on:
Does this problem resonate with you?
What's missing for it to be useful in your workflow?
Any obvious gaps I'm not seeing?
Would genuinely appreciate thoughts from builders and ai power users who've using Claude code, Codex or Opencode Cli, for multiple parallel sessions at the same time.
And If you are interested to testing, you can go to www.denker.ai for more details.
Replies