sarah

i sold my bootstrapped tech start-up and now i am building @Composio while getting my masters in AI @UTAustin

hiring

[sarah @sarahfim]

Despite being told no, I'm open-sourcing TrustClaw.

You can now deploy a production-ready personal agent service with over 1000+ app integrations in a single command, straight to @vercel with npx @composio/trustclaw deploy

I was inspired by @openclaw to build a simple web app where anyone could create their own 24/7 personal assistant and connect it to Gmail, Google Calendar, Notion, Slack, GitHub, HubSpot, Linear… well everything, and securely through OAuth/sandbox execution.

It went viral on X, reached over a thousand users in less than 48h, and revenue began pouring in.

If you are thinking like a company, you'd probably keep that locked up. But why should I be the reason you spend another year scrolling instead of building?

So today, I'm open-sourcing TrustClaw anyway.

> 24/7 agents that act across Gmail, Notion, GitHub, Slack, Linear, Jira, and 1000+ apps
> OAuth and sandboxed execution, so users don't have to hand agents passwords or raw API keys
> Supports multiple users and authentication right outside of the box with @better_auth

Repo is open, MIT licensed.

If I were starting an AI company today, I'd clone this, pick a market, and begin shipping with Claude Code.

Honestly so excited to see what comes out of this.

[Composio @composio]

We're excited to be at AI Engineer World's Fair 2026. 🚀

If you're building AI agents, developer tools, or enterprise AI systems, come meet the Composio team. We'll be talking about secure integrations, real-world tool use, and what it takes to move agents from demos to production.

Looking forward to connecting with builders. See you there!

#AIEngineering #AIAgents #AgenticAI #DeveloperTools #LLMs #AIEngineerWorldsFair

[sarah @sarahfim]

@shawnbuilds @cursor_ai @composio @vercel @supermemory so happy to see content that goes step by step into building such an incredible agent from 0 to 1, great watch!!

[sarah @sarahfim]

@nikos1 bro get out of our office 😭😭😭

[Maya Lekhi @maya_l39]

first week at @vercel! stoked to be working in the office of the cto with @andrewqu

[sarah @sarahfim]

@maya_l39 @vercel @andrewqu lfg!!

[sarah @sarahfim]

@juliafedorin @databricks exposing enterprises😭😭

[sarah @sarahfim]

@dopabees @whop LETSSGOO!!

[sarah @sarahfim]

let’s goooo!!!!

[Guillermo Rauch @rauchg]

One of our most requested features, longer Vercel function runtime, is here.

What looks like an innocent tweak of a constant… is actually the conclusion a multi-year compute platform investment.

Builds, Sandbox, and now Functions run on our homegrown microVM-based Fluid compute infrastructure.

This investment has enabled innovations like function multi-concurrency, Active CPU pricing, and Secure Compute for private connectivity to existing cloud workloads… with lots more to come, in fact, in the coming days and weeks 😎

[sarah @sarahfim]

time for @composio to spread like jam

(@wespreadjam btw)

[jia @jia_seed]

onboarding @composio to jam!

which office next?

[sarah @sarahfim]

@jia_seed @composio @juliafedorin @wespreadjam i was blown away!! so excited to continue jamming :))

[Nikos @nikos1]

Thanks for all the love everyone :)

Got tons of inbound from potential customers and teammates, investors, and cool people.

This post also taught me a lot about my positioning. I had some great chats w people on it and have already pivoted language a lot to distance Glen from some adjacent products.

Kicking off design partnerships this week and next week, excited to share more as Glen shapes up.

I will invite more people to use Glen from the waitlist in the coming weeks, keep an eye on your emails.

[sarah @sarahfim]

@rifago no what YOUR doing means a lot

[sarah @sarahfim]

the world made a silent trade: hand over your data, get intelligence back

back in 2021 when i was researching privacy-preserving AI, the problem was obvious: you couldn't run inference privately in the cloud. the leaks that followed surprised no one paying attention.

running models locally is amazing, but for companies that need scale, it is not an answer. the only real fix is securing AI at the cloud infrastructure level which was expensive and completely infeasible... until now.

that's exactly what @rifago is building at @ZimaLabsIO, and after months of using it I'm blown away: no latency hit, fully scalable, mathematically secure.

this will be the default for how secure AI runs in the cloud.

thank you for your work team! excited for whats next :)

[rifa @rifago]

your llm chat history is probably on the web.

in the past 18 months:

- major AI lab leaked millions of chat logs due to a misconfigured database
- thousands of Grok conversations showed up in google results

this is why i'm building Zima, a new way to chat and access LLMs securely where no one ever sees or can see your data, including us.

Zima hosts open-source models on this infrastructure and retains zero data, so whatever you say to your llm stays private, forever.

your personal chats with AI should never be up for sale.

try Zima at the link below :)

[sarah @sarahfim]

@juliafedorin @contextconor this looks insane, excited!!

[sarah @sarahfim]

@tuanle madara has nothing on @tuanle

[sarah @sarahfim]

@shriyalola @browserbase for FREE?? and i get to meet my idol @shriyalola ?? i'm there.

[sarah @sarahfim]

see you @browserbase :) https://t.co/YSOwplO0S0

[sarah @sarahfim]

@rifago i have been waiting for the day we could finally run inference in the cloud securely, and it is finally here!! this is incredible @rifago thank you for your work.

[sarah @sarahfim]

@SammyTourani @composio @try_glen @nikos1 hello @nikos1 nice to meet you

[sarah @sarahfim]

@shriyalola the agents yearn for the browser