Launching today
HQ
The company brain your AI runs on
13 followers
The company brain your AI runs on
13 followers
Today everyone uses AI their own way. HQ gets your whole team working as one. A workspace that makes your AI smarter the more you use it, and turns one person's wins into the whole team's default. You get sharper. They get faster.
Hey Product Hunt ā Corey here, founder of HQ.
I run sixteen companies day-to-day. Every one of them needs me, my team, and our AI tools to share the same context ā what we decided last Tuesday, which customer owns which thread, the policy we set six months ago that nobody remembers. For two years I rebuilt the same scratch system over and over: a folder of files that Claude could read, a shared way to give every new session the same starting point, a way for my team to all work from the same playbook.
HQ is that system, productized. It's a local workspace your team's AI tools all read from ā Claude Code, Cursor, Codex, whatever's next. Knowledge, skills, policies, and per-company workers live in one place and stay in sync across everyone's machines. Every AI session starts informed instead of starting cold.
A few things to know:
ā¢ Local-first and open source. The free starter kit is MIT-licensed and runs entirely on your machine. Get it at hqforwork.com. HQ Pro adds an encrypted vault and team sync on top of the same files you already own.
ā¢ Agent-agnostic. HQ is the layer underneath. Use whatever AI you trust above it ā we don't compete with the model, we compete with the cold start.
ā¢ Scales without changing shape. Solo founders, small teams, and enterprises all use the same primitives. Engineers at Adobe and a handful of other teams are running it daily.
A few things I'd love feedback on today:
1. Where does the cold-start problem hurt you most ā code, customer, or company-knowledge work?
2. If you've tried building this kind of shared context yourself, what broke first?
3. What's the one feature that would make you put HQ in front of your whole team this week?
I'll be in the comments all day. AMA.
Hi everyone, I'm of the engineers building HQ.
Before HQ, my day was a row of terminal tabs, each running an AI agent that knew nothing about the others. When a teammate's agent learned something the hard way, mine made the same mistake the next day.
HQ fixed that for us. Our whole team and every agent we run work from one shared company brain. I now run a few teams of agents, one fixing production errors and one shipping releases, and I check on all of them from one place.
Question for anyone running more than one agent at a time: how do you keep track of what each one is doing? Tabs, a board, sticky notes?