Former Anthropic and DeepMind lead Behnam Neyshabur launches Mirendil with $200 million to automate frontier AI engineering

VIEWS2.9KBOOKMARKS7

https://mirendil.com/careers.html

We are fortunate to have following advisors, investors, angels, and supporters: @istoica05 @ramin_m_h @pearvc @seedtosunflower @neo @aixventureshq @naval @assaf_rappaport @realSharonZhou @bahat_merav @mirrokni @prfsanjeevarora @dmdohan @Yuhu_ai_ @Azaliamirh @DLeitersdorf @danintheory @mobav0 @imkelvinxu @FeinbergVlad @ananyaku @djstrouse @MikowaiA @actuallyHadi @mo_norouzi @lachygroom @MaikaThoughts The Trustees of the UPenn & Diane Greene

2h2.9K197

LIKES24

Azalia Mirhoseini@Azaliamirh

Congrats on the launch! Super excited to see what @bneyshabur and the talented @MirendilAI team build to accelerate science!

Behnam Neyshabur@bneyshabur

Today, I’m excited to formally announce @MirendilAI with my amazing co-founders Harsh Mehta, Shayan Salehian, and Tara Rezaei!

We’re fortunate to work with @a16z and @kleinerperkins, who led our seed round of $200M, followed by a major investment from NVIDIA, among others.

Mirendil exists to accelerate science and technology, and through them, to help solve humanity's most pressing problems.

Self-accelerating AI R&D is the most direct path to delivering on AI's broader promise, which is why we believe the most important application of AI is AI itself. Get this loop right, and it compounds. It fundamentally changes the rate of progress itself across all domains.

We believe this capability should be democratized. It should be used to power all scientific efforts trying to innovate at the frontier. There are far more important problems—and broader ones—than any single lab can take on, so more groups should be able to pursue them.

This pulls concentration of power away from a few labs: businesses and science labs can own their AI and infrastructure, keep their margins, and control their own destiny instead of ceding it all to a single AI lab.

We’re a small team with a singular focus. Our founding team consists of 20 researchers and engineers from frontier institutions including Anthropic, xAI, Google DeepMind, and OpenAI, united by a passion for science and a drive to build the technologies that move it faster. If you want to build the system that builds systems, join us!

@HarshMeh1a, @shayan_, @tararezaeikh

1h1.6K243

RETWEETS16

a16z@a16z

We're thrilled to lead Mirendil's $200M seed round.

Frontier AI work has been locked inside a few big labs.

@MirendilAI is building a system that can help anyone do AI work: they train frontier models that are expert at AI R&D and build the product around it. The result is a system that loops over research and engineering problems on its own, making progress without human intervention.

It's like a coding agent built for AI research that controls its own GPUs.

Mirendil is one of the few teams with the experience and the priors to make the end-to-end system work. @bneyshabur, @HarshMeh1a, @shayan_, and @tararezaeikh come out of Anthropic, Google DeepMind, and xAI.

Welcome to the a16z family, Mirendil.

By @BornsteinMatt and @MaikaThoughts

Behnam Neyshabur@bneyshabur

Today, I’m excited to formally announce @MirendilAI with my amazing co-founders Harsh Mehta, Shayan Salehian, and Tara Rezaei!

We’re fortunate to work with @a16z and @kleinerperkins, who led our seed round of $200M, followed by a major investment from NVIDIA, among others.

Mirendil exists to accelerate science and technology, and through them, to help solve humanity's most pressing problems.

Self-accelerating AI R&D is the most direct path to delivering on AI's broader promise, which is why we believe the most important application of AI is AI itself. Get this loop right, and it compounds. It fundamentally changes the rate of progress itself across all domains.

We believe this capability should be democratized. It should be used to power all scientific efforts trying to innovate at the frontier. There are far more important problems—and broader ones—than any single lab can take on, so more groups should be able to pursue them.

This pulls concentration of power away from a few labs: businesses and science labs can own their AI and infrastructure, keep their margins, and control their own destiny instead of ceding it all to a single AI lab.

We’re a small team with a singular focus. Our founding team consists of 20 researchers and engineers from frontier institutions including Anthropic, xAI, Google DeepMind, and OpenAI, united by a passion for science and a drive to build the technologies that move it faster. If you want to build the system that builds systems, join us!

@HarshMeh1a, @shayan_, @tararezaeikh

1h42K217119

REPLIES2

Andrew Curran@AndrewCurran_

Today has been an extremely busy news day. How about a Recursive Self Improvement startup to top it off.

1h886170

Mamoon Hamid@mamoonha

We're at a breakthrough moment in AI. The most direct path to its promise is AI that accelerates its own research, and @MirendilAI is building exactly that. A frontier lab rebuilt from scratch around AI R&D.

This team has made mind blowing progress in the short few months they’ve been working together using their own systems. @bneyshabur, @HarshMeh1a @shayan_, @tararezaeikh and the Mirendil team have a rare mix of research depth, scale experience, and the instinct to ship. Expect(ing) big things. We’re proud to have been there from day one.

https://www.kleinerperkins.com/perspectives/mirendil-building-the-system-that-builds-systems/

32m1.3K155

Behnam Neyshabur@bneyshabur

We are fortunate to have following advisors, investors, angels, and supporters: @istoica05 @ramin_m_h @pearvc @seedtosunflower @neo @aixventureshq @naval @assaf_rappaport @realSharonZhou @bahat_merav @mirrokni @prfsanjeevarora @dmdohan @Yuhu_ai_ @Azaliamirh @DLeitersdorf @danintheory @mobav0 @imkelvinxu @FeinbergVlad @ananyaku @djstrouse @MikowaiA @actuallyHadi @mo_norouzi @lachygroom @MaikaThoughts The Trustees of the UPenn & Diane Greene

2h1.4K144

Mamoon Hamid@mamoonha

We're at a breakthrough moment in AI. The most direct path to its promise is AI that accelerates its own research, and @MirendilAI is building exactly that. A frontier lab rebuilt from scratch around AI R&D.

This team has made mind blowing progress in the short few months they’ve been working together using their own systems. @bneyshabur, @HarshMeh1a @shayan_, Tara and the Mirendil team have a rare mix of research depth, scale experience, and the instinct to ship. Expect(ing) big things. We’re proud to have been there from day one.

https://www.wsj.com/tech/ai/anthropic-veterans-startup-seeks-to-help-scientists-develop-their-own-ai-09e2f3e5?mod=tech_lead_story

1h988182

Behnam Neyshabur@bneyshabur

https://www.wsj.com/tech/ai/anthropic-veterans-startup-seeks-to-help-scientists-develop-their-own-ai-09e2f3e5?mod=author_content_page_1_pos_1

Behnam Neyshabur@bneyshabur

https://mirendil.com/careers.html

2h2.4K202

Preetum Nakkiran@PreetumNakkiran

excited about our mission! there are many areas of science which AI can almost-certainly help accelerate, but which currently would require many human-hours of expert work (eg, AlphaFold). Reducing this barrier, ideally to 0 expert-hours, is IMO both tractable and impactful.

Behnam Neyshabur@bneyshabur

Today, I’m excited to formally announce @MirendilAI with my amazing co-founders Harsh Mehta, Shayan Salehian, and Tara Rezaei!

We’re fortunate to work with @a16z and @kleinerperkins, who led our seed round of $200M, followed by a major investment from NVIDIA, among others.

Mirendil exists to accelerate science and technology, and through them, to help solve humanity's most pressing problems.

Self-accelerating AI R&D is the most direct path to delivering on AI's broader promise, which is why we believe the most important application of AI is AI itself. Get this loop right, and it compounds. It fundamentally changes the rate of progress itself across all domains.

We believe this capability should be democratized. It should be used to power all scientific efforts trying to innovate at the frontier. There are far more important problems—and broader ones—than any single lab can take on, so more groups should be able to pursue them.

This pulls concentration of power away from a few labs: businesses and science labs can own their AI and infrastructure, keep their margins, and control their own destiny instead of ceding it all to a single AI lab.

We’re a small team with a singular focus. Our founding team consists of 20 researchers and engineers from frontier institutions including Anthropic, xAI, Google DeepMind, and OpenAI, united by a passion for science and a drive to build the technologies that move it faster. If you want to build the system that builds systems, join us!

@HarshMeh1a, @shayan_, @tararezaeikh

1h1.6K241

Andrew Curran@AndrewCurran_

They are hiring:

Harsh Mehta@HarshMeh1a

I'm excited to announce what we've been working on for the past few months: @MirendilAI

Our goal is to broadly accelerate science and technology with AI — from solving grand challenges to helping engineers and scientists in their day-to-day work at the lowest levels of detail.

We're at an unusual moment in history where progress in a single technology can massively accelerate every field of science and engineering simultaneously, and we think there's no higher-leverage way to spend our time than making it happen faster.

We're a small team in SF, and we're hiring: https://mirendil.com/careers.html

1h1.8K51

Alex Konrad@alexrkonrad

we've finally run out of Lord Of The Rings startup names, so ChatGPT is generating new ones

53m42661

Andrew Curran@AndrewCurran_

https://www.wsj.com/tech/ai/anthropic-veterans-startup-seeks-to-help-scientists-develop-their-own-ai-09e2f3e5

Andrew Curran@AndrewCurran_

1h72880

Andrew Curran@AndrewCurran_

Arno@aarnogau

It's been a long time coming. I left xAI a little over a month ago now.

It was a pleasure to work with incredible people turning a new lab into a force to be reckoned with. I am incredibly grateful and proud of the opportunity to lead the Grok web team from no product impacting hundreds of millions of users. Also grateful for helping kick start the Grok Build project which has turned into something truly amazing. I learned a ton from watching the product team grow from 6 to 30 people from scaling every service on fire to caring about the minute and millisecond details for users.

Towards of the end of my time at xAI, I realized that model capabilities are the true bottleneck for building out better AI products. I helped out the code modeling team on our full stack and long horizon experts but couldnt help but think that maybe the real problem was the approach to modeling.

I'll be joining Behnam, Harsh, Tara, and Shayan at @MirendilAI to work on the next generation of AI products.

As model training and hosting costs decrease, and they will decrease a lot, we will see a renaissance in how humans interact with the world. Models that do real work means people can reinvest their time into ideas and creativity.

Of course, we arent there yet, so that's what we're building at Mirendil.

Accelerating research has revolutionary immediate use cases in enabling more GPU architectures with kernel perf to allowing anyone to improve models for their own needs. Theres a lot more underserved industries and even personal use cases that can be improved with more discovery.

If youre interested in acceleration dm me or apply here! https://mirendil.com/careers.html

1h76330

Mo Bavarian@mobav0

Honored to back this team!! 🚀

Behnam Neyshabur@bneyshabur

Today, I’m excited to formally announce @MirendilAI with my amazing co-founders Harsh Mehta, Shayan Salehian, and Tara Rezaei!

We’re fortunate to work with @a16z and @kleinerperkins, who led our seed round of $200M, followed by a major investment from NVIDIA, among others.

Mirendil exists to accelerate science and technology, and through them, to help solve humanity's most pressing problems.

Self-accelerating AI R&D is the most direct path to delivering on AI's broader promise, which is why we believe the most important application of AI is AI itself. Get this loop right, and it compounds. It fundamentally changes the rate of progress itself across all domains.

We believe this capability should be democratized. It should be used to power all scientific efforts trying to innovate at the frontier. There are far more important problems—and broader ones—than any single lab can take on, so more groups should be able to pursue them.

This pulls concentration of power away from a few labs: businesses and science labs can own their AI and infrastructure, keep their margins, and control their own destiny instead of ceding it all to a single AI lab.

We’re a small team with a singular focus. Our founding team consists of 20 researchers and engineers from frontier institutions including Anthropic, xAI, Google DeepMind, and OpenAI, united by a passion for science and a drive to build the technologies that move it faster. If you want to build the system that builds systems, join us!

@HarshMeh1a, @shayan_, @tararezaeikh

1h55850

hattie@oh_that_hat

Mirendil has a fantastic team — excited to see what they build!

Behnam Neyshabur@bneyshabur

Today, I’m excited to formally announce @MirendilAI with my amazing co-founders Harsh Mehta, Shayan Salehian, and Tara Rezaei!

We’re fortunate to work with @a16z and @kleinerperkins, who led our seed round of $200M, followed by a major investment from NVIDIA, among others.

Mirendil exists to accelerate science and technology, and through them, to help solve humanity's most pressing problems.

Self-accelerating AI R&D is the most direct path to delivering on AI's broader promise, which is why we believe the most important application of AI is AI itself. Get this loop right, and it compounds. It fundamentally changes the rate of progress itself across all domains.

We believe this capability should be democratized. It should be used to power all scientific efforts trying to innovate at the frontier. There are far more important problems—and broader ones—than any single lab can take on, so more groups should be able to pursue them.

This pulls concentration of power away from a few labs: businesses and science labs can own their AI and infrastructure, keep their margins, and control their own destiny instead of ceding it all to a single AI lab.

We’re a small team with a singular focus. Our founding team consists of 20 researchers and engineers from frontier institutions including Anthropic, xAI, Google DeepMind, and OpenAI, united by a passion for science and a drive to build the technologies that move it faster. If you want to build the system that builds systems, join us!

@HarshMeh1a, @shayan_, @tararezaeikh

57m21020

Andrew Curran@AndrewCurran_

Mirendil@MirendilAI

We are proud to formally announce Mirendil. Our mission is to accelerate science and technology through democratizing frontier AI R&D.

We’re a small team with a singular focus. Our founding team consists of 20 researchers and engineers from frontier institutions including Anthropic, xAI, Google DeepMind, and OpenAI, united by a passion for science and a drive to build the technologies that move it faster. If you want to build the system that builds systems, join us!

We’re fortunate to work with @a16z and @kleinerperkins, who led our seed round of $200M, followed by a major investment from NVIDIA, among others.

1h1.4K30

Arno@aarnogau

@bneyshabur @MirendilAI @a16z @kleinerperkins Marginal cost of training goes down 👉 AI for everyone

2h3364

Mirendil@MirendilAI

We are proud to formally announce Mirendil. Our mission is to accelerate science and technology through democratizing frontier AI R&D.

We’re a small team with a singular focus. Our founding team consists of 20 researchers and engineers from frontier institutions including Anthropic, xAI, Google DeepMind, and OpenAI, united by a passion for science and a drive to build the technologies that move it faster. If you want to build the system that builds systems, join us!

We’re fortunate to work with @a16z and @kleinerperkins, who led our seed round of $200M, followed by a major investment from NVIDIA, among others.

Behnam Neyshabur@bneyshabur

Today, I’m excited to formally announce @MirendilAI with my amazing co-founders Harsh Mehta, Shayan Salehian, and Tara Rezaei!

We’re fortunate to work with @a16z and @kleinerperkins, who led our seed round of $200M, followed by a major investment from NVIDIA, among others.

Mirendil exists to accelerate science and technology, and through them, to help solve humanity's most pressing problems.

Self-accelerating AI R&D is the most direct path to delivering on AI's broader promise, which is why we believe the most important application of AI is AI itself. Get this loop right, and it compounds. It fundamentally changes the rate of progress itself across all domains.

We believe this capability should be democratized. It should be used to power all scientific efforts trying to innovate at the frontier. There are far more important problems—and broader ones—than any single lab can take on, so more groups should be able to pursue them.

This pulls concentration of power away from a few labs: businesses and science labs can own their AI and infrastructure, keep their margins, and control their own destiny instead of ceding it all to a single AI lab.

We’re a small team with a singular focus. Our founding team consists of 20 researchers and engineers from frontier institutions including Anthropic, xAI, Google DeepMind, and OpenAI, united by a passion for science and a drive to build the technologies that move it faster. If you want to build the system that builds systems, join us!

@HarshMeh1a, @shayan_, @tararezaeikh

2h6.9K3410

Kiana Ehsani@ehsanik

Amazing team 😍 Excited to see what @bneyshabur and his crew build!

Behnam Neyshabur@bneyshabur

Today, I’m excited to formally announce @MirendilAI with my amazing co-founders Harsh Mehta, Shayan Salehian, and Tara Rezaei!

We’re fortunate to work with @a16z and @kleinerperkins, who led our seed round of $200M, followed by a major investment from NVIDIA, among others.

Mirendil exists to accelerate science and technology, and through them, to help solve humanity's most pressing problems.

Self-accelerating AI R&D is the most direct path to delivering on AI's broader promise, which is why we believe the most important application of AI is AI itself. Get this loop right, and it compounds. It fundamentally changes the rate of progress itself across all domains.

We believe this capability should be democratized. It should be used to power all scientific efforts trying to innovate at the frontier. There are far more important problems—and broader ones—than any single lab can take on, so more groups should be able to pursue them.

This pulls concentration of power away from a few labs: businesses and science labs can own their AI and infrastructure, keep their margins, and control their own destiny instead of ceding it all to a single AI lab.

We’re a small team with a singular focus. Our founding team consists of 20 researchers and engineers from frontier institutions including Anthropic, xAI, Google DeepMind, and OpenAI, united by a passion for science and a drive to build the technologies that move it faster. If you want to build the system that builds systems, join us!

@HarshMeh1a, @shayan_, @tararezaeikh

30m21440

rohan anil@_arohan_

@bneyshabur @MirendilAI @a16z @kleinerperkins Congrats bro! 😎 💪💪

Behnam Neyshabur@bneyshabur

Today, I’m excited to formally announce @MirendilAI with my amazing co-founders Harsh Mehta, Shayan Salehian, and Tara Rezaei!

We’re fortunate to work with @a16z and @kleinerperkins, who led our seed round of $200M, followed by a major investment from NVIDIA, among others.

Mirendil exists to accelerate science and technology, and through them, to help solve humanity's most pressing problems.

Self-accelerating AI R&D is the most direct path to delivering on AI's broader promise, which is why we believe the most important application of AI is AI itself. Get this loop right, and it compounds. It fundamentally changes the rate of progress itself across all domains.

We believe this capability should be democratized. It should be used to power all scientific efforts trying to innovate at the frontier. There are far more important problems—and broader ones—than any single lab can take on, so more groups should be able to pursue them.

This pulls concentration of power away from a few labs: businesses and science labs can own their AI and infrastructure, keep their margins, and control their own destiny instead of ceding it all to a single AI lab.

We’re a small team with a singular focus. Our founding team consists of 20 researchers and engineers from frontier institutions including Anthropic, xAI, Google DeepMind, and OpenAI, united by a passion for science and a drive to build the technologies that move it faster. If you want to build the system that builds systems, join us!

@HarshMeh1a, @shayan_, @tararezaeikh

1h61530

Former Anthropic and DeepMind lead Behnam Neyshabur launches Mirendil with $200 million to automate frontier AI engineering

Story Overview

Founders bring deep bench from four frontier labs

No product or benchmarks are public yet