An AI hire for one number you care about. It works daily in Gmail, GitHub, Search Console and the rest, asks your team before anything goes out, and gets a 1:1 every two weeks to improve. Open source. Runs free on the ChatGPT or Grok plan you already pay for.
Looper is an open-source app for running AI teammates, called loopers. Each looper owns one business metric, and the metric sets its unit of work:
Organic clicks: an article
First response time: a support ticket
Unreconciled transactions: a bank transaction
Each unit moves through stages. An article goes New → Drafted → Approved → Done: the looper writes it, opens a pull request, and asks you to approve the merge. On Fridays it holds a retro. It compares the week’s work with the change in clicks, scores the week 1 to 5, and adds next week’s experiments to its kanban.
How this differs from tools you may use:
Grok and ChatGPT are organized around a conversation. You ask, they answer or act, and you choose the next step. Zapier and n8n are organized around a workflow. You define the trigger and the steps. Notion, Trello and Miro are organized around pages and boards that people update.
Looper is organized around units of work. The looper creates them, moves them through stages, and keeps four views up to date:
Notion-style database: one row per article
Trello-style kanban: its tasks and the ones waiting on you
Apple-style calendar: when it runs
Miro-style canvas: its plan, references and notes on what worked
Three ways to run it:
Self-host: MIT license, one-click deploy to Vercel and Neon.
Hosted, free: connect your ChatGPT or Grok subscription. Web search and computer use are pay as you go.
Hosted with credits: $49/mo, including Claude credits.
Which metric would you start with, and what would its unit of work be?
No comment highlights available yet. Please check back later!
About Looper on Product Hunt
“If Grok Bot supported team collaboration and was Open Source”
Looper launched on Product Hunt on November 2nd, 2026 and earned 96 upvotes and 1 comments, placing #14 on the daily leaderboard. An AI hire for one number you care about. It works daily in Gmail, GitHub, Search Console and the rest, asks your team before anything goes out, and gets a 1:1 every two weeks to improve. Open source. Runs free on the ChatGPT or Grok plan you already pay for.
Looper was featured in Productivity (665.1k followers), Task Management (84.3k followers), Artificial Intelligence (483.2k followers) and GitHub (41.5k followers) on Product Hunt. Together, these topics include over 359.7k products, making this a competitive space to launch in.
Who hunted Looper?
Looper was hunted by Pavel Kucherbaev. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.
Want to see how Looper stacked up against nearby launches in real time? Check out the live launch dashboard for upvote speed charts, proximity comparisons, and more analytics.
Pavel Kucherbaev
Hi Product Hunt, I’m Pavel.
Looper is an open-source app for running AI teammates, called loopers. Each looper owns one business metric, and the metric sets its unit of work:
Organic clicks: an article
First response time: a support ticket
Unreconciled transactions: a bank transaction
Each unit moves through stages. An article goes New → Drafted → Approved → Done: the looper writes it, opens a pull request, and asks you to approve the merge. On Fridays it holds a retro. It compares the week’s work with the change in clicks, scores the week 1 to 5, and adds next week’s experiments to its kanban.
How this differs from tools you may use:
Grok and ChatGPT are organized around a conversation. You ask, they answer or act, and you choose the next step. Zapier and n8n are organized around a workflow. You define the trigger and the steps. Notion, Trello and Miro are organized around pages and boards that people update.
Looper is organized around units of work. The looper creates them, moves them through stages, and keeps four views up to date:
Notion-style database: one row per article
Trello-style kanban: its tasks and the ones waiting on you
Apple-style calendar: when it runs
Miro-style canvas: its plan, references and notes on what worked
Three ways to run it:
Self-host: MIT license, one-click deploy to Vercel and Neon.
Hosted, free: connect your ChatGPT or Grok subscription. Web search and computer use are pay as you go.
Hosted with credits: $49/mo, including Claude credits.
Which metric would you start with, and what would its unit of work be?