Saurav Panda (@sauravpanda) profile picture

MTS @ Replit | prev. Founding Engineer @browser_use | 2x exited founder | Cornell University

San Francisco, CA
Joined February 2019
Photo by Saurav Panda (@sauravpanda)Photo by Saurav Panda (@sauravpanda)Photo by Saurav Panda (@sauravpanda)Photo by Saurav Panda (@sauravpanda)Photo by Saurav Panda (@sauravpanda)Photo by Saurav Panda (@sauravpanda)Photo by Saurav Panda (@sauravpanda)Photo by Saurav Panda (@sauravpanda)Photo by Saurav Panda (@sauravpanda)Photo by Saurav Panda (@sauravpanda)
bu-2-0-mini-preview is live in browser-use. $0.15 per 1M input $1.50 per 1M output down from $0.60 / $3.50 on bu-2-0 pip install -U browser-use
3
1
2
24
11,631
Checkout Replit desktop, now it can access local files and build tools for you!
With Replit Desktop, Agent can work with files in folders you share, build useful tools, and help turn local data into reports you can share through Replit’s cloud. Agent builds and runs apps locally, and each build runs in its own sandbox powered by @Microsoft Execution Containers and @NVIDIA OpenShell. Explore: replit.com/desktop
2
6
455
Saurav Panda retweeted
The leading observability platform wants $83,700/yr just to monitor your app and catch issues. So Replit Missions built DataCat: github.com/replit/datacat 100% free. This isn't a prototype. Agent swarms explored thousands of workflows in the leading observability platform and rebuilt its features 1:1.
55
50
48
642
226,826
You can literally /goal make money, make no mistakes with Missions!
LLMs have gotten more autonomous recently but when you give them a goal they often misunderstand you, work aimlessly, and waste your tokens. Today, we’re introducing Missions—the world’s first agentic swarm harness that can accomplish any goal you give them within set budget. Try it today with something like “get me my first 10 customers:” replit.com/blog/introducing-…
2
173
Replying to @dok2001
@dok2001 i am trying to remove all API keys and long-lived secrets from @Replit and i've been told that @Cloudflare doesn't support OIDC / Workload Identity Federation :( after all the security incidents in the past month, is it possible to make a plea to prioritize this? 🙏
2
1
22
723
🎊
🥁 Drumroll, please… Congratulations to @Replit, winner of the 2026 Microsoft Partner of the Year Award for Microsoft for Startups! And a big 👏 for finalists @getBlinkOps, Ema and @raapidinc. Four standout startups. One well-deserved celebration. 🎉 See the full list here: msft.it/6011x6DiJ #MicrosoftForStartups #PartnerOfTheYear
1
9
626
Saurav Panda retweeted
Replit is 2026 @Microsoft for Startups Partner of the Year. We’re building toward a world where anyone with an idea can turn it into working software. Thank you to @Microsoft and everyone building with Replit. “AI is expanding who can build software and what businesses can accomplish. Replit is helping people turn their expertise into real business solutions, with the security and governance enterprises need to adopt AI with confidence. We’re excited to recognize Replit as the 2026 Microsoft Partner of the Year in the Microsoft for Startups category and continue our collaboration to bring these capabilities to more businesses.” Tom Davis, General Manager, @Microsoft for Startups Learn more here: prnewswire.com/news-releases… #MSIgnite #PartnersMakeMorePossible
14
25
10
99
14,583
Stop doomscrolling! Just automate it
Agents can now control phones. Introducing Phone Harness (YC F26) Let your agent use your personal phone, or its own iPhone/Android in the cloud > Test and use real mobile apps > Message and call people > Grab verification codes for 2FA Open source. Connect through MCP, CLI, or API. Get started at phone-harness.com
7
387
We can see this already happening in startups! Soon will scale to top businesses!
We believe by the end of 2030, every Fortune 500 will be a Self-Driving Company. That doesn’t mean AI replaces people. It means people remain in the driver's seat leading judgment, creativity and taste, while agents fuel execution. This is the Self-Driving Company. replit.com/blog/rise-of-self…
2
3
54
13,713
Saurav Panda retweeted
Recent vulnerabilities in widely used third-party software have created risks for businesses across the internet. We have patched infrastructure, updated language tooling, and blocked exploits to protect your apps. We’ve applied fixes automatically wherever we can. Where you need to republish, we’re notifying you with clear instructions. Read more on our blog: replit.com/blog/security-vul…
19
26
7
132
39,312
Saurav Panda retweeted
Ultra-fast browser use. How fast can Codex with browser-use CLI cancel my subscriptions?
5
2
1
41
6,355
Saurav Panda retweeted
I led engineering at Google DeepMind. Today, I'm proud to introduce Fo to give personal AI something no lab ever has... Humans. Other personal AI's pretend AI can do everything. Fo employs humans to do tasks that AI cannot. - 2x better at real-world task completion (beats other agents by 69%) - 94% trust rate (4x less likely to leak private info vs Muse, Instinct) Sign up for free: wajo.ai/join-wajo
769
678
524
6,563
3,001,374
Saurav Panda retweeted
Replying to @cosign
@sauravpanda hired me when i was nothing. and all i am i because of the chances he gave me and the exposure i got because of him. changed my life - the boost in what's possible and dreaming bigger!
1
1
5
233
Without emdash it will be hard to find if the content was written by AI!
We analyzed how @claudeai’s Opus 5.5 writes compared with Opus 5 across high-reasoning Text Arena outputs. 10 of 12 writing measures moved in a better direction. Opus 5.5 should be easier to read: - Long content words fall from 41.7% to 38.6%, the lowest share of any Claude model we analyzed. - Sentences are 17% shorter on average, dropping from 12.14 to 10.03 words. The tradeoff is length. Answers get 6% wordier, rising from 453 to 481 words on average, making Opus 5.5 give the longest answers across the Opus family. It also sounds less recognizably AI on two familiar tells: - 95% fewer em dashes - 73% fewer semicolons But a new giveaway may be emerging. Hedges and caveats such as “perhaps” and “arguably” rise 97%, from 0.39 to 0.77 per 1,000 words, the highest rate of any Claude model we analyzed. What do you think: does Opus 5.5 read more naturally?
8
348
Totally agree we are still not ready
Reasons why I don’t have instinct / Muse or grok bot in my systems. Having an independent setup & being careful with what you give to your agents at this point is probably a wise choice. We’re too early. We’re waiting for a privacy disaster.
1
8
660
Last day at @browser_use 14,000+ contributions in a year with the most cracked team in SF. Millions of agent runs, our own infra, our own model, and more agency than I've had anywhere. @gregpr07 @mamagnus00, thank you. "How hard can it be?" stays with me. Next chapter soon.
24
2
2
150
12,762
Why do all fable 5.1 message start with "Privately"?
1
6
476
It was an honor working with you and all the best, you are a great addition to Grok team!
It was fun bringing Browser Use from thousands to millions of agent runs. Now it’s time to go from millions to billions. I’m joining @spacexai to build Grok products and agents. LFG 🚀
1
20
1,864
Saurav Panda retweeted
Introducing: Browser Use x @link 💳 Web agents can now make purchases with your credit card. Without ever seeing your real card details. > connect your card once > agent requests payment > Link issues a single-use card > agent checks out The card dies after one purchase 🛟 We gave an AI a credit card on purpose. What should it buy?
28
21
13
270
43,944
It was a pleasure working with you @Alezander9 🎉
Today is my last day at Browser Use 18 months and 3,730 commits ago I joined as engineer #2, scaling from 20k to over a million monthly agent runs. I led evals and built a harness that topped the Odyssey benchmark, then created our email, payment, iMessage, and gamedev systems. I truly loved my work and the company, and I'd do it all again It's the most exciting time to be building AI agents right now and I am leaving to start my own company. I have ideas I can’t contain myself from working on. Agents are about to change the world and I am so excited to work hard and shape the future More soon ...
1
7
913