A SOTA agent, Droid, topping terminal-bench scores against other agentic coding tools. Use OpenAI/Claude/Google/BYO models Accessible in your browser, Slack, Linear, mobile, CLI and IDE with instant session sync. Use Droid SDK to build agents on top of Droid.
Hi Product Hunt, I’m Eno, co-founder & CTO at Factory. Most AI dev tools make you pick sides: switch IDEs, accept one model, run in their cloud. Builders need choice and control. Factory is a platform for AI coding agents that works wherever you do.
The same agent shows up on the web, in your IDE/terminal and CLI, and inside tools like Slack and Linear. You keep your stack. Run in our managed sandbox or on your own machine. Set autonomy to your comfort level. See transparent reasoning and native diffs so nothing ships without your say. It scales to million‑line repos. And you choose the best model per task: Claude, GPT‑5, Gemini, with no lock‑in.
Try pasting a specification and come back to a clean PR. Fan out parallel refactors and migrations from the CLI while CI stays green. Tag the agent in Slack or assign a ticket and get a one‑shot resolution. If you want speed without giving up control, try Factory. I’m here all day for questions and workflow requests!
Having gone from playing with prompts to GPTs, to Claude Projects, to IDEs w/ agents + multiple keys - I can safely say the latter is orders of magnitude better for anyone who is leaning in fast.
So to have that same capability but across web, IDE/terminal, and Slack is malto bueno. Impressed.
Been using Factory AI at my company for a few months, and am already impressed! Better than other AI coding tools, much better understanding of existing code and infra, and less time spent fixing bugs or anti patterns.
This looks really cool! I’m impressed by the real, time collaboration tools. How does the workflow management feature help teams keep track of their tasks?
I've seen so many posts about teams struggling with communication and missing deadlines. It sounds like you’re addressing a real pain point.
If you want to connect with teams that need this kind of help, I’ve seen conversations about these issues. It could really benefit from engaging with those discussions.
I've used Codex and Claude Code, this is a legit alternative that doesn't lock you into any single company's models. When Gemini 3 releases, I have no doubt that if it's very good, Factory will integrate.
Factory feels snappy and is currently (as I write this on Sep 25, 2025) sitting at the top of the Terminal-Bench leaderboard: https://www.tbench.ai/leaderboard
I prefer using GPT 5 with Factory only because Claude's models are very expensive and sometimes slower.
I'd also recommend a tool like RepoPrompt to help with finding and feeding appropriate context to any AI coding tool you use (whether Codex, CC, Factory, etc). It'll help you avoid burning through all of your credits
I'm not associated with any the companies or products mentioned above, just a dev.
The OG @bentossell is back and what a product too. Ben gave me early access to Droid CLI and it built out the whole of herenow.fyi from scratch in a day, having it launched in prod and running on my site by the evening.
Was a super experience, the best I've had so far in terms of planning and building without me being hands on, definitely worth a look if you're into Claude Code, Codex etc.
Excited to see where the team take it!
This could seriously help our dev team cut down on endless context switching.
Factory allows me to switch from IDE to Slack to mobile without missing a beat...... instant session sync is such a lifesaver when juggling projects on the go. Super slick work, and congrats for the launch!
Excited to be back on Product Hunt 10 years after I first started my career there in 2015. Since then I’ve built, launched, sold, and invested… but I haven’t hunted a product in over a decade. Today we’re launching Factory CLI, part of our mission to bring autonomy to software engineering. It’s the agent-native terminal tool I wish I had when I was first stumbling through dependencies, linters, and GitHub.
Feels full circle: from Product Hunt intern to hunting again as Factory’s new Head of DevRel.
About Factory on Product Hunt
“Agent-native development on Web, IDE, CLI, Slack and mobile”
Factory launched on Product Hunt on September 25th, 2025 and earned 178 upvotes and 20 comments, placing #11 on the daily leaderboard. A SOTA agent, Droid, topping terminal-bench scores against other agentic coding tools. Use OpenAI/Claude/Google/BYO models Accessible in your browser, Slack, Linear, mobile, CLI and IDE with instant session sync. Use Droid SDK to build agents on top of Droid.
Factory was featured in Artificial Intelligence (466.2k followers), Development (5.8k followers) and SDK (738 followers) on Product Hunt. Together, these topics include over 90.2k products, making this a competitive space to launch in.
Who hunted Factory?
Factory was hunted by Ben Tossell. 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 Factory stacked up against nearby launches in real time? Check out the live launch dashboard for upvote speed charts, proximity comparisons, and more analytics.
Ben Tossell
Hi Product Hunt, I’m Eno, co-founder & CTO at Factory. Most AI dev tools make you pick sides: switch IDEs, accept one model, run in their cloud. Builders need choice and control. Factory is a platform for AI coding agents that works wherever you do.
The same agent shows up on the web, in your IDE/terminal and CLI, and inside tools like Slack and Linear. You keep your stack. Run in our managed sandbox or on your own machine. Set autonomy to your comfort level. See transparent reasoning and native diffs so nothing ships without your say. It scales to million‑line repos. And you choose the best model per task: Claude, GPT‑5, Gemini, with no lock‑in.
Try pasting a specification and come back to a clean PR. Fan out parallel refactors and migrations from the CLI while CI stays green. Tag the agent in Slack or assign a ticket and get a one‑shot resolution. If you want speed without giving up control, try Factory. I’m here all day for questions and workflow requests!