Building verification layer @PramaanaLabs
Palo Alto, CA
Joined November 2021
- Tweets604
- Following403
- Followers445
- Likes15.8K
Reminds me of Shaka Laka Boom Boom
en.wikipedia.org/wiki/Shaka_…
Join us and help us r̵e̵d̵u̵c̵e̵ remove hallucinations from AI
Hiring across Product, Design and Engineering. DM for details
jobs.ashbyhq.com/pramaana-la…
After 12 months of deep research, we know our thesis works.
We set out to build AI that takes ownership to give verified answers; not just through evals/benchmarks, but with audit-trails of mathematical proof for correctness.
After a year of careful research, we've hit critical mass. Now we take it to production, in domains where the stakes are high, and being wrong is not an option. This is impeccable timing, because the hard science is ready to ship, and yet early enough that the people who join now will define and shape how it ships.
The problems are still unsolved, and the bar is high.
If you're the person who gets restless when the answer is just "good enough," we would love to talk to you.
Check out our open roles here: jobs.ashbyhq.com/pramaana-la…
Mit Dave retweeted
483 days ago, @5aitec and I raised $20M from @binnybansal to build ShopOS
Millions of brand images and videos later we learned: the model was never the hard part. Getting it right the 10,000th time is
Today, we’re launching @sloosh_ai to give every creator that secret sauce 👇
Hardware acquired ✅ Meta opened Muse ✅
time to build 🔥
teapawt.org/mitdave/status/2103195…
🚨 side project alert 🚨
Announcing Muse Gadgets, an open source ESP32 firmware and Linux sdk so that you can make hardware devices that work with Muse.
Grab an API token from gadgets.muse.ai and point your favorite coding agent at the github repo to build your own peripherals for Muse.
When your stack nails something it wasn’t even built to do - and somehow makes it look easy and fast⚡️⚡️⚡️
@PramaanaLabs prover, Hardy, took a quick break from verifying mission-critical reasoning across tax, legal, and financial domains last weekend to tackle something a bit more recreational: IMO 2026.
Hardy solved 6/6 problems in 7 hours using Lean 4. Our autoformalization model, Panini, translated the natural language problem statements into Lean 4.
Full write-up & GitHub repo with Lean proofs: pramaanalabs.ai/blog/pramaan…
Appreciation post (and feature request) for @claudeai Design
@PramaanaLabs team has been shipping presentations and marketing assets with our rudimentary design system. Work that previously required an idea person, designer, and frontend engineer can now be owned end to end by one person.
Chat and Pro mode also make it surprisingly easy for non-frontend and frontend teammates to collaborate on the same artifact. Genuinely great.
[Feature request] A better way to publish these assets as web apps on a custom domain.
Sending them to @vercel works, but it creates a new project each time. It would be much more useful if Claude Design could update an existing Vercel project or, even better, publish directly.
Git-backed project would be the cherry on top.
@ClaudeDevs @trq212
Mit Dave retweeted
Current Frontier LLM models hallucinate with Tax queries on phase-outs, miss quadratic interactions, and leave money on the table; all while sounding confident.
Formal verification of these outputs at run-time is the only meaningful way to address this.
Cell-by-cell 1040 modeling + machine-checked proofs that your tax plan is compliant and provably optimal. Audit-defensible. Fearless.
Authored by: Yoshiki Takashima
Read how we turn complex tax optimization into verifiable truth → pramaanalabs.ai/blog/audit-d…
Lots of work ahead, but I'm super excited to build the product around our research at @PramaanaLabs
We just raised $27M led by @khoslaventures (with Accel, Boldcap, Nexus, Premji Invest & Unbound) to take AI from probably right to provably right.
Focusing on turning formal verification + Lean into real, production-grade systems. Can't wait for what's next.
Removing "…can make mistakes. Check important info." disclaimer from AI generated answers.
Actively hiring the best talent. DM if this is your problem too.
Say no to hallucinations!
Today, I'm thrilled to announce Pramaana's $27M seed, led by @khoslaventures.
The foundational domains that hold the world together: tax, law, finance, healthcare; all run on certainty. Probabilistic AI can't give them that. We’ve been asked to accept wrong answers with AI as ‘hallucinations’, while in traditional software terms, it’s just a bug. And a wrong answer in such mission-critical domains is more than just a bug, it's a liability that could have catastrophic impact.
We built Pramaana to deliver a 100% trustable experience to the domains that run on certainty: AI that is provably correct, not probabilistically correct. We turn statute and regulation into machine-verifiable code, so every output ships with mathematical proof of correctness. Our mission is to make AI take ownership of it’s work.
Pramaana in Sanskrit stands for “means of valid knowledge”, and we’re going to achieve that by formalizing the world’s knowledge.
Removing ambiguity from AI-generated answers.
Join us on June 10th to learn more
luma.com/597otp0a?tk=HucuKq
Join the room where AI stops saying 'Sorry' !
Hear from the industry's best minds speak on the problem and possibilities of verification in AI.
It's time to verify AI !
@khoslaventures @PramaanaLabs @boldcapfund
Mit Dave retweeted
The room I wanted to be in when we started Pramaana didn't exist. The people working on formal verification weren't in the same room as the people deploying AI into regulated domains.
One year later, we're building that very room.
June 10th, San Francisco.
verificationsummit.ai
Outputs that sound right are not the same as outputs we can prove right.
The next phase of AI runs on verified systems: formal methods, theorem provers, runtime checks, provable agents.
It's time to verify AI !
Join us on the 10th of June at SF !
@PramaanaLabs @boldcapfund
Google IO showcased AI glasses and the live demo was really cool. Specially the watch companion app to quickly glance the answers and ability to capture images and modify it using Nano Banana Pro. Updated WearableAI to have live activities, a watch companion app and image generation via Nano Banana Pro (try asking it to capture a photo and do edits on it) 😃
Try it on your RayBan Meta's - testflight.apple.com/join/Gs…
