Launching our AI workflow tool on Product Hunt Thursday. Here is what we built (and one thing I keep going back and forth on)
Almost every team I talk to has a process they can describe in one sentence but have never actually built. "When a lead comes in, qualify it, update the CRM, and ping the right person." Everyone knows what should happen. Almost nobody has it running, because wiring the tools, the data, and the AI together turns into its own project every time.
That gap is what we spent the last two years on. ModuleX lets you describe a process in plain words and get back a real workflow you can run.
The decision I keep chewing on: we made the output an editable graph instead of a black box. You can open any step, see what it does, and change it by hand. It would have been faster to ship a "just trust the agent" experience, but every team we talked to wanted to see and control what runs. Curious if other builders here have felt that same pull between magic and control.
A few specifics for this crowd:
- Prompt to a running workflow, then run it from chat, API, or SDK
- Tools, data, and AI agents in one graph you can read and edit
- 140+ integrations, all OAuth-verified
- Free tier is 100 runs a month, no card
Sezer (my co-founder) built the whole thing. He previously took Thinkbuddy to around 20K users.
We go live on Product Hunt this Thursday. I will share the link here when we do. If you have launched on PH before, I would genuinely take any last-minute advice. What is the one thing you wish you had done differently on your launch day?
the pull is real but i think you answered it right for this category. workflow tools are the one place 'just trust the agent' fails hard, because the cost of a silent wrong step is somebody's CRM getting mangled and nobody noticing for a week. people forgive a black box for image gen, they don't forgive it for anything touching their data. the editable graph is what makes it a tool people put real processes on instead of a toy they test once. i'd lean into the control imo, the people who want pure magic were never going to trust it with a real process anyway.
What stood out to me is that "magic" and "control" may not actually be opposite choices.
The harder question might be which part of the workflow users are willing to trust before they demand visibility.
Getting that wrong can make the same product feel either effortless or uncomfortable.