300,000+ @hyperagentapp sessions on Browserbase since April.
Airtable built a platform where anyone can deploy a fleet of always-on agents.
Hyperagent gives you the agents. Browserbase gives them the web.
Agent browser use is still so underrated.
If your agents can't actually see webpages, they're walking around blindfolded. So much high-value work cannot be done through APIs alone.
Hyperagent runs @browserbase to visually navigate the web live.
300,000+ @hyperagentapp sessions on Browserbase since April.
Airtable built a platform where anyone can deploy a fleet of always-on agents.
Hyperagent gives you the agents. Browserbase gives them the web.
Doors just opened for day 3 of @aiDotEngineer World’s Fair
Swing by our booth (next to OpenAI) for a demo of our new Agent product and exclusive Navigate merch 🎁
@pk_iv will be speaking at 11:40am about the future of agents on the web (Floor 2, Room 2024)
Computer use is finally reliable. It just needed a real harness. Now there is one.
Starting today, anyone can ship a SOTA browser agent, batteries included.
Introducing Browserbase Agents: one prompt and one API call is all you need to automate the whole web.
Swing by P5 at the @aiDotEngineer World's Fair to hang with the team and see what we're building 👋
→ 6 talks, including a keynote from @pk_iv
→ mini-podcast recording live at our booth
→ cool merch + something new 👀
→ sponsoring browser dogs tomorrow (6/30)
Derek built bb, the internal agent the whole company runs on. 100% feature-request coverage. 99% of support answered in under 24 hours.
@derekmeegan went on @insecureagents to break down how it works, and how it stays secure enough to trust.
Out now ⬇️
100% feature-request coverage. 99% of support answered in under 24 hours. Built by a few engineers.
@derekmeegan built bb, the internal agent serving the entire team at @browserbase. His advice: if you want to scale an agent's reach make sure the harness is verifiably secure.
Gemini 3.5 Flash + CUA is impressive: on our benchmarks, it performs similarly to frontier models while being on average ~80% cheaper.
It's a reliable model for powering browser agents that balance speed and accuracy on real-world tasks.
Couldn’t have said it better ourselves ❤️
Patient data still lives behind dozens of pre-API portals. So we built agents that don’t just fetch data, they reason.
Hundreds of browser agents run per case on @browserbase, turning 60-minute manual reviews into 60-second clinical
Magicare turned a 60-minute admissions decision into 60 seconds with Browserbase.
Patient data lives behind dozens of portals built before APIs existed. The only way in is the way a human gets in: through a browser.
So @magicareai runs hundreds of browser agents per case on
Magicare turned a 60-minute admissions decision into 60 seconds with Browserbase.
Patient data lives behind dozens of portals built before APIs existed. The only way in is the way a human gets in: through a browser.
So @magicareai runs hundreds of browser agents per case on
1/n
For browser agents, a major bottleneck in evaluation is truthful scoring on the live web. A task is only as good as your ability to confirm the agent actually did it, on a real site whose state keeps moving and that the agent can potentially misreport.
So we took matters
