Sanjay Ganapathy

48 posts

Sanjay Ganapathy

Sanjay Ganapathy

@sanjaygsub

Co-Founder @ Pramaana Labs | ex-Google Deepmind ASI's search for Truth

San Francisco Katılım Ekim 2022
62 Takip Edilen685 Takipçiler
Sabitlenmiş Tweet
Sanjay Ganapathy
Sanjay Ganapathy@sanjaygsub·
Nine years at Google, three on Gemini at DeepMind, working on post-training with people I'd follow anywhere. Close enough to scale to see how far it goes, and where it stops. Today's AI is remarkable, but jagged: it hallucinates fluently. In high-stakes domains like healthcare, legal, or finance, a confident mistake is worthless. Right as I was wrestling with this, @krishnan_rag , @ranjan_vittal and I converged on a wild idea: AI that proves its own work. Proofs, not disclaimers. So I left. Nine months later, at Pramaana Labs, we're building it: a cracked team training foundation models to formalize human knowledge on a scale hitherto undreamt of, and bring formally verified AI to the real world. A future where AI is provably correct, not just probably.
ranjan_raj@ranjan_vittal

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.

English
13
36
345
60.8K
Sanjay Ganapathy
Sanjay Ganapathy@sanjaygsub·
Sridhar Vembu received the distinguished alum award at IIT Madras in 2016, the year I graduated. I distinctly remember seeing him for the first time, giving a talk on entrepreneurship to our graduating CS batch. A decade on, to hear him give a shout out to @PramaanaLabs and our thesis on national TV is definitely a goosebumps moment! Thank you @svembu 🙏 Let's bring efficient, verified intelligence to the world!
Republic@republic

#RepublicSummit2026 |Sridhar Vembu (@svembu), Co-founder & Chief Scientist, Zoho sheds light on what happens to jobs as software productivity keeps increasing. From promise to proof. From aspiration to execution. #RepublicSummit2026 brings together the biggest names in power, policy and progress to discuss India's emergence as a global shaping force. Nation First. Always. WATCH LIVE: youtube.com/live/c_OEQhm9G… #GreatPowerIndia #RepublicSummit2026 #NationFirst #ArnabGoswami Presented By: RP-Sanjiv Goenka Group (@rpsggroup)| Co-Presented By: ZOHO (@Zoho) & TVS Motor Company(@tvsmotorcompany) | Powered By: Sister Nivedita University(@snuindia) | Co-Powered By: Kalyani (@kssldefence) & Bharat Forge Ltd (@BharatForgeLtd), Samtel Avionics (@SamtelAvionics), Ravin Group(@Ravingroup) and MeghaShrey(@sseemasinghh), Ease My Trip (@EaseMyTrip) | State Partners: UP Govt(@UPGovt), Haryana Govt(@DiprHaryana), Andhra Pradesh Govt(@IPR_AP), Uttarakhand Govt(@ukcmo) and Assam Govt | Housing Partner: Gaurs Group(@Gaurs_Official) | Prayer Partner: Cycle Agarbatti | Skincare Partner: Dr Rashel(@drrashelindia) | Knowledge Partner: Shardha University(@sharda_uni) | Stainless Partner: Jindal Stainless(@Jindal_Official) | Vision Partner: Alcon | Special Partners: Sanskriti University, Gallantt Group (@Gallantt_Group), Pan Bahar(@PanBaharElaichi), Krishna's Ayurveda(@HerbalKrishna), BackBay, Wagh Bakri(@waghbakri), Ferns N Petals, Nandini and Mysore Sandals | Celebration Partner: Allied blenders and distillers(@ABDL_India)

English
5
55
276
76K
Sanjay Ganapathy
Sanjay Ganapathy@sanjaygsub·
I am incredibly grateful for @vkhosla's continued support of this vision. His energy is unmatched, and the absolute conviction he brings to the table leaves me in total awe every time!
ranjan_raj@ranjan_vittal

Last week @vkhosla anchored our summit. Injured, when stepping back would've been expected, he showed up and turned it into an example of how verification can transform healthcare. The conviction he brings to every founder, no matter how small the company, is humbling. The GOAT.

English
1
2
13
1.2K
Sanjay Ganapathy retweetledi
Vinod Khosla
Vinod Khosla@vkhosla·
Auto formalization will be an important new area.
ranjan_raj@ranjan_vittal

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.

English
15
23
200
80.5K
Noam Shazeer
Noam Shazeer@NoamShazeer·
I’m excited to share that I’ll be joining OpenAI and look forward to working with the exceptional team there. It was a difficult decision to move on. I’m incredibly proud of the amazing team at Google and everything we’ve built together. It has been an honor and a pleasure to work with all of you.
English
981
866
16.2K
9.4M
Sanjay Ganapathy
Sanjay Ganapathy@sanjaygsub·
@jnptl This provides mathematical proof for AI outputs in domains where humans use deduction based on a set of objective rules
English
0
1
5
340
jaimin patel
jaimin patel@jnptl·
@sanjaygsub Its really good launch video. Would you be able to explain what is the actual product in simple terms? and where in the life cycle of AI it fits.
English
1
0
0
469
Sanjay Ganapathy
Sanjay Ganapathy@sanjaygsub·
Nine years at Google, three on Gemini at DeepMind, working on post-training with people I'd follow anywhere. Close enough to scale to see how far it goes, and where it stops. Today's AI is remarkable, but jagged: it hallucinates fluently. In high-stakes domains like healthcare, legal, or finance, a confident mistake is worthless. Right as I was wrestling with this, @krishnan_rag , @ranjan_vittal and I converged on a wild idea: AI that proves its own work. Proofs, not disclaimers. So I left. Nine months later, at Pramaana Labs, we're building it: a cracked team training foundation models to formalize human knowledge on a scale hitherto undreamt of, and bring formally verified AI to the real world. A future where AI is provably correct, not just probably.
ranjan_raj@ranjan_vittal

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.

English
13
36
345
60.8K
Sanjay Ganapathy
Sanjay Ganapathy@sanjaygsub·
@AbhishekEswaran Yes that is why we formalize the knowledge first in Lean. We will be sharing a technical blog on this soon
English
0
1
1
383
Abhishek Eswaran
Abhishek Eswaran@AbhishekEswaran·
@sanjaygsub interesting! we built an agent to file taxes and it messed up by by choosing iran instead of india in the dropdown! would love to learn how you do it
English
1
0
0
607
Sanjay Ganapathy
Sanjay Ganapathy@sanjaygsub·
I have known @krishnan_rag since high school when we were preparing for Olympiads. Coincidentally we even got the same score on IIT-JEE! I've worked alongside some ridiculously brilliant folks at DeepMind but Krishnan remains one of the absolute sharpest minds I know. Stoked to be working alongside him at @PramaanaLabs 🚀
English
0
0
16
736
Sanjay Ganapathy retweetledi
Perplexity Fund
Perplexity Fund@PerplexityFund·
Not many launch videos are this inspiring-- feeling like the future is getting brighter thanks to @ranjan_vittal and team's formal AI verification for real-world domains. Congrats to @PramaanaLabs on the seed raise! Proud to back such a visionary team.
ranjan_raj@ranjan_vittal

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.

English
0
3
10
2.3K
Sanjay Ganapathy retweetledi
Accel in India
Accel in India@AccelIndia·
In tax, law, finance, government, and healthcare, AI still cannot guarantee its answers are correct. A doctor still reads the diagnosis. A lawyer still checks the contract. A tax accountant still signs the return. Not because AI cannot produce an answer, but because when it is wrong in a high-stakes domain, it cannot be held responsible. That is AI's accountability gap. And @PramaanaLabs is building the fix. Pramaana applies formal verification to these domains at scale, converting complex knowledge like tax codes, clinical protocols, and financial regulations into a formal language that machines can reason over with mathematical certainty. The system either returns a machine-checkable proof that an answer is correct or shows exactly where the reasoning breaks. If it cannot prove an answer, it will not provide one. What makes this particularly compelling is the team's proximity to the problem. Ranjan Rajagopalan (@ranjan_vittal), Krishnan Raghavan (@krishnan_rag), and Sanjay Ganapathy Subramaniam spent years building AI systems at Google Maps, Glean, and Google DeepMind, confronting firsthand the challenges of accuracy, reliability, and trust that they are now setting out to solve. Partnering with Pramaana Labs to take AI from probably right to provably right. @prashanthp • @anagh_prasad • #AccelFamily
Accel in India tweet media
English
3
5
21
3.1K
ranjan_raj
ranjan_raj@ranjan_vittal·
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.
English
99
154
1.3K
317.3K
Sanjay Ganapathy retweetledi
ranjan_raj
ranjan_raj@ranjan_vittal·
Reflecting on our event - @fv_summit Every moment of this event on the 10th of June, has been a deeply humbling experience for me. Right from our event being over-subscribed on Luma for it's very first edition, to sitting across from Vinod for the fireside, to seeing 200+ people fill up the venue despite being a working day, to having marquee speakers on stage sharing a common vision, to the deeply technical and profound discussions at the panel, to the passionate networking conversations going on late till 11pm. All of these have been very reassuring for us, and we're committed to further the cause of Verification in AI. Thanks to @vkhosla for his dedication to the event and setting the tone for the evening. Thanks to every speaker who flew in, and thanks to every attendee who could join us. Thank you to my co-founders, our team at @PramaanaLabs; @khoslaventures and @BoldCap for being in this with us. This is just the beginning. It's time to verify AI.
ranjan_raj tweet mediaranjan_raj tweet mediaranjan_raj tweet mediaranjan_raj tweet media
English
0
4
11
557
Sanjay Ganapathy
Sanjay Ganapathy@sanjaygsub·
Last week I joined a panel on the frontiers of AI × Verification at the @fv_summit, with @evelovesolive, @diagram_chaser, @DjDvij and @sathyanellore . From researching the jagged frontier of Gemini's capabilities, one thing was clear: Verifiability is the bottleneck to superhuman AI on any class of tasks. And verification needs rigorous specification. The catch: rigorous specification doesn't come naturally to us. We run on intuition — and intuition can generate an answer, but it can't certify one. Yet certainty is exactly what high-stakes work needs to stay reliable and compound. And frontier models won't learn that rigor from human data. At @PramaanaLabs, we're teaching it directly: pairing AI with a formal specification language and training on machine-checkable rewards.
English
0
3
11
27.6K
Sanjay Ganapathy retweetledi
Nikesh Arora
Nikesh Arora@nikesharora·
The frontier model problem is a breadth versus depth problem. Consumer needs breadth, the wider your aperture, the more relevant your model. Breadth suffers from a false positive problem, it ranges between 10-30%, with clever prompting and checks you can hit the lower end of the range, but it tends to be free or subsidized. Consumers still seemm to be satisfied, and consumption growing! However, Frontier Models harvest the usage data to inform future models. On the other hand. The enterprise wants depth, their tolerance for error is low, this needs more context data, training and harnesses and guardrails is high. The frontier models aren't ready yet to provide that, hence the FDEs, and solutions consultants who build that capacity for every enterprise. But enterprise is the only route ATM to build a sustainable economic model. The risk, consumer losses mount. Enterprise value accrues to solution providers. In the meanwhile, models are aggressively pursuing Enterprise profit pools, while solution providers are building orchestrators to arbitrage token pricing. So there's many a push and a pull in the equation. If will be an epic battle, my instinct tells me, value could accrue to the application and proprietary data layers. Will be fun to watch.
Chamath Palihapitiya@chamath

Game theory from here is super interesting: Original Mags (Google, Amazon, Microsoft, Meta) now have a serious non-zero opportunity to tank the frontier labs. Go to the government, kneecap the labs’ motion of putting the latest models out in the wild, become the trusted gatekeeper between the labs and the public at large (including internationally) by having the labs go through their clouds (AWS, GCP, Azure) and implement strict KYC to seal the deal. The frontier labs should have seen this coming years ago and implemented a robust KYC for just this moment. The fact they didn’t is kind of concerning. Why did they not do it? Best guess is because it would have changed the run-rate revenues (downward) which would have then changed funding dynamics - lower valuations, more dilution, less secondary. A valuation reset may happen now anyways, except the labs may end up with less control and more restrictions at the end of it. At the same time, everyone is already clamoring about token prices of the old models from the labs anyways… This couldn’t be a better setup for open source and neoclouds. Big question is can they meet the moment? There are too few of them and their progress seems sporadic at best.

English
44
62
544
192.5K
Sanjay Ganapathy retweetledi
Marco Pavone
Marco Pavone@drmapavone·
I look forward to participating in the Verification Summit (verificationsummit.ai) and sharing my perspective on Physical AI safety. I strongly agree that verification and validation are key frontiers for unlocking Physical AI in high-stakes, high-reliability applications, from autonomous cars to industrial robotics! @fv_summit @khoslaventures @PramaanaLabs @boldcapfund
Vinod Khosla@vkhosla

I think this will turn out to be a very important area. Founders should work on things AI is not good at.

English
0
7
25
4.3K
Sanjay Ganapathy retweetledi
The FV Summit
The FV Summit@fv_summit·
5 days to go ! 14 speakers. One stage. One question: Can we hold AI accountable where it matters the most? Join the room where AI stops saying 'Sorry'! It's time to verify AI ! @khoslaventures @PramaanaLabs @boldcapfund
The FV Summit tweet media
English
1
5
12
880

Keşfet

@PramaanaLabs @svembu @vkhosla @NoamShazeer @jnptl @krishnan_rag @ranjan_vittal @AbhishekEswaran