it seems like agentic commerce is finally solved with stripe link. and we're excited to partner with stripe to enable all agents to transact online. your agent can now natively make payments on the web using link cli + browserbase. one shot prompt:

we built blazing fast computer/browser use with Jev + @Stagehanddev. this task cost $0.001 and executed at near instant speed (in a remote browser btw) the loop: observe the page, send a11y tree as state + actions as questions, Jev decides the next action, then Stagehand executes it.

Hosting a panel this Friday: Pacing the Frontier? AI is moving fast. The question is how fast should it move, and what are we building toward? Joining me: @pk_iv, Browserbase @NaderLikeLadder, NVIDIA @FurqanR, Nebula @trq212, Anthropic Limited spots. RSVP below.

LangChain's CEO says they built their gateway a year too late, and it's still costing them. I sat down with Harrison Chase, Co-Founder & CEO of LangChain, to talk about what's changed since LangChain started as a side project in 2022, why sub-agents rarely work the way people expect, and why LangChain now runs its own hiring, content, and coding work through internal agents in Slack. 00:00 Why "I use agents for everything" is a red flag in an interview 01:21 How LangChain started as a fall 2022 side project 05:18 From LangSmith to LangGraph: rebuilding as the ground kept shifting 09:06 The gateway LangChain built a year too late 13:23 Code mode: why agents writing code beats human-style tools 19:04 Should the agent run inside the sandbox, or call it as a tool? 28:31 The agent identity problem: whose credentials does it use? 45:43 Running LangChain's own GTM, content, and coding agents in Slack 48:34 The AI slop problem, and why they built a Slack emoji for it The full pod is also available on YouTube, Apple Podcasts, or Spotify.

LangChain's CEO says they built their gateway a year too late, and it's still costing them. I sat down with Harrison Chase, Co-Founder & CEO of LangChain, to talk about what's changed since LangChain started as a side project in 2022, why sub-agents rarely work the way people expect, and why LangChain now runs its own hiring, content, and coding work through internal agents in Slack. 00:00 Why "I use agents for everything" is a red flag in an interview 01:21 How LangChain started as a fall 2022 side project 05:18 From LangSmith to LangGraph: rebuilding as the ground kept shifting 09:06 The gateway LangChain built a year too late 13:23 Code mode: why agents writing code beats human-style tools 19:04 Should the agent run inside the sandbox, or call it as a tool? 28:31 The agent identity problem: whose credentials does it use? 45:43 Running LangChain's own GTM, content, and coding agents in Slack 48:34 The AI slop problem, and why they built a Slack emoji for it The full pod is also available on YouTube, Apple Podcasts, or Spotify.

this is the most important talk i've given in my time as a founder. in the next twelve months, we believe that every task that humans do online will be assisted by agent. watch this if you want to see the future that we're building for @browserbase

i built Scout, a small but mighty deep research agent that lives in slack. it uses @browserbase search and fetch + proxies to reach corners of the internet agents couldn't access before.

this is the most important talk i've given in my time as a founder. in the next twelve months, we believe that every task that humans do online will be assisted by agent. watch this if you want to see the future that we're building for @browserbase

I trained a fly to use a browser. A simulated circuit built from part of the fly connectome makes the decisions. Stagehand turns them into browser actions. Check it out at

yesterday we organized our first conference in @browserbase history. having our entire team together on that stage was one of the most meaningful moments of our journey so far. thank you to all who attended and I can't wait for NAVIGATE 2027

your portal to another universe opens at 10am in downtown SF. introducing NAVIGATE by @browserbase - one day of content for the best teams building on the frontier of intelligence.

Introducing our Browser Agent Evals. We benchmark frontier and open models using various harnesses on computer-use tasks. Our evals are fully open source and reproducible via our Evals CLI.

We just made Astra's computer use 2.5x faster with Stagehand. Astra works by executing code against the a11y tree, we built a translator that turns Astra's playwright commands into @Stagehanddev. Astra chose to batch the entire logo into one command and one-shotted it.

we are now approaching the "second order problems" phase of agents browsing the web. ex: how do we prevent abuse? who is responsible for this agent's actions? what access should it have? thankfully, i believe these are all solvable problems. exciting times ahead!

GPT-6 Astra broke the internet with its ability to use blender and generate 3D worlds, all a byproduct of it's new computer use capabilities. I dove into the source code and wrote about the loop that makes Astra so good:

Founders used to manually log in every day to send and receive payments because their bank had no API. That changed with Mercury. Founder Immad Akhund (@immad) built the bank he wished existed after hitting that wall himself. What started as banking for startups now includes agentic cards, letting an AI agent hold a virtual card and spend on your behalf. He joined me on Navigators to break down how Mercury decides what to build next, and where he draws the line on what AI should be trusted to do without a human clicking send. 00:00 Why banks never built the API he needed 02:34 A trillion-dollar category still run by incumbents 05:29 "I don't like jumping on hype trains" 07:34 Agentic cards: letting AI spend on your behalf 09:15 AI as a junior employee you've never met 11:09 Why fintech is 3x harder than software 36:39 The "cosmic cube": Deciding what to build next 37:27 The two-year mistake of waiting on the credit card Listen to our full conversation on Navigators, available on YouTube, Spotify, and Apple Podcasts.

We got early access to Fable 5.1 and evaluated it extensively across our internal benchmarks. It outperformed all other frontier models (including Fable 5) on accuracy scoring 92.11%, while being 60% cheaper than Fable 5 per task.