Bryce, the CUDA Colonel

Member of Technical Staff at @modal. Led prog langs @NVIDIA 2017-2026. @adspthepodcast cohost. C++ Lib Evolution chair emeritus. Frequent flyer. Horology fan

[Bryce, the CUDA Colonel @blelbach]

The latest revision of @INCITS/@isostandards COBOL comes out this year

The goals of COBOL sound normal today:

- Portable
- Freely available
- Designed by the community

In 1959 it was radical & unprecedented

It was also conceived of & led by women

This is the story of COBOL

[Bryce, the CUDA Colonel @blelbach]

@venkat_systems @ekzhang1 @modal Not sure I'm in the same caliber of hire as @ekzhang1 but appreciated!

[Bryce, the CUDA Colonel @blelbach]

uberkernel

[Daniel Kim @learnwdaniel]

What’s bigger than a mega kernel? Omegakernel?

[Modal @modal]

Compute shouldn’t be the bottleneck for your next big idea.

We're launching Modal for Builders, a program that provides compute credits to builders working on ambitious AI projects.

Have a demo you’ve been wanting to build? A research question you want to explore? Or another idea that needs compute?

We’d like to hear about it. Apply with your idea and an estimate of the compute you’ll need. Projects may also be featured on Modal’s social channels.

Some projects we’ve backed in the past:

[Bryce, the CUDA Colonel @blelbach]

@homuraakem92126 @TrungTPhan I think you may have confused Bill Daly with Ian Buck.

[ADSP: The Podcast @adspthepodcast]

📢 ADSP Episode 311 is out! 📢 In this episode, @blelbach and @code_report chat with Zach Laine about his career, the career changes he has made (including leaving NVIDIA), GPUs and more! https://t.co/dbCoEirOkJ

[Bryce, the CUDA Colonel @blelbach]

@ekzhang1 @modal @thinkymachines Very exciting that I get to sit across from this guy!

[Eric Zhang @ekzhang1]

I’m back at @modal

Working at @thinkymachines was really something. I’ll miss my coworkers, and a lot has happened since I joined last year - all sorts of ML and production systems, witnessing what goes into a model, throwing new kernels at training and inference, scaling research infra, making interaction models fast and good, wrangling RL dynamics.

It was also my second job, so perhaps a formative experience. When I joined, I discovered I was the youngest at the company (like 60 people) and felt pretty out-of-place, surrounded by ex-OpenAI/Meta/GDM/Mistral folks. Anyway, the company reinvented itself many times, and I think with the recent model releases we can see it’s on a good path.

Thanks to @cHHillee for originally recruiting me to Thinky and my boss and various coworkers (whom I won’t dox) for teaching me very much

I saw this sunset rainbow the other day in NYC, and it’s the only time aside from July 4 I’ve seen everyone on the streets stop and stare at the sky for a bit, together. It was brief. This won’t last forever.

Anyway, hope to have more for you soon… much to build! The Modal team is very strong. I’m glad to be here.

[Tony Mongkolsmai @tonymongkolsmai]

Exciting thing here (really exciting!). Introducing our new Nsight Perf SDK for AI, a beta preview, that enables a very, very low overhead collection technology at profiling data rates aka 10Khz+ with the ability to attach to running workloads, see your CUDA kernel traces and tie those to hardware metrics...

Check it out, give feedback and yes we are building other fun stuff on top of this... stay tuned!

https://t.co/GvTNLAUA5y

[Akshat Bubna @akshat_b]

This model is going to be a big deal. We ran K3.1 through our evals for our internal SRE agent - it uses more tokens, but the quality improvement is shockingly good and well worth the cost.

[Benۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗۗ☁️ @benswerd]

November 2027 sandbox tier list.

Anybody not on here is either based or deeply deeply mid.

[Modal @modal]

Kimi K3.1 Pro from @Kimi_Moonshot is now available on Modal.

A major post-training update to Kimi K3, with expanded reinforcement learning for stronger coding, financial research, and long-horizon agentic work.

Kimi K3.1 Pro supports text and images with 1M context.

[Simran Arora @simran_s_arora]

Introducing the Together Compiler: continual compilation for AI inference! I’m super proud to finally share this work from our research team! ✨

Today’s inference is “Polymorphic” or shape-shifting, it doesn’t sit still. On OpenRouter, 80% of traffic goes to models released in the last three months and the top ten most popular open models each use a different architecture. Our customers constantly change their usage behaviors. Together Compiler leverages agents to continually optimize our deployments immediately as these changes occur.

Together Compiler research highlights include:
1. 252x better throughput: TKCC lets agents try 252x more optimization experiments per hour compared to prior agent inference optimization systems.
2. State-of-the-art TPS across InferenceX workloads we tried: TKCC outperforms the best InferenceX engine across all tested settings in just 24 hours & 2 nodes per compile. Including 1111 TPS on MiniMax M3.1 B200 and 1205 TPS on DeepSeek v4.1 Flash GB300 at batch size 1. NVIDIA, AMD, Trainium2.
3. Continuous recompilation: TKCC currently serves many trillions of production tokens per day and adapts to traffic shifts within just a few hours of recompilation.
4. Bitter lessons: TKCC’s design builds on 2+ years of building frameworks to accelerate AI Inference, starting from ThunderKittens and Megakernels. Compilers have always used abstractions and IRs, but we show that agentic compilers now reach higher performance ceilings without abstractions.

From an incredible team @reyna_abhyankar @dylan__lim @realDanFu @togethercompute!!

[Harry Stebbings @HarryStebbings]

Why confidence was the only thing that held me back from starting a company; now it is worth $15.7BN

“I never had quite enough confidence in myself. A lot of what was missing was just the confidence.

I saw this big opportunity in the market and thought, ‘Why is no one else doing this? It is such an obvious idea.’ That led me to think maybe I should start a company.

Five years later, here we are. It has gone pretty well. In hindsight, maybe I should have started much earlier.” @bernhardsson

Love to hear your thoughts @sarahcat21 @eladgil @swyx @mattturck

[Nick Khami @skeptrune]

a single GPU now makes more than the federal minimum wage of $7.25/hr

fucked up to say, but isn't that a sign of compute being underpriced?

one B300 produces much more value than a minimum wage human ☹️

[Bryce, the CUDA Colonel @blelbach]

@lauriewired Wait you sing?

[Bryce, the CUDA Colonel @blelbach]

I love Toronto but it's not even the best food city in Canada. Montreal is the best food city in the world.

[Sacha Sayan @sachasayan]

Btw, after having been to like a hundred countries now: Toronto is the best food city in the world and it's not even close.

[Chris Lattner @clattner_llvm]

Agentic coding benefits from powerful abstractions, just like humans do.

[Bryce, the CUDA Colonel @blelbach]

@charles_irl @modal @nvidia Charles you do not want my tiny Polish girlfriend stalking you down for hair revenge!

[Bryce, the CUDA Colonel @blelbach]

The REAL reasons I joined @modal:

- Cool logo in my Twitter handle.
- Compatible with my green wardrobe from @nvidia.
- Unlimited GPU credits and tokens.
- @charles_irl's hair.

[ADSP: The Podcast @adspthepodcast]

📢 ADSP Episode 310 is out! 📢 In this episode, @blelbach and @code_report chat about Bryce leaving @nvidia, where he is headed next and more! 🥳 https://t.co/9geMXot8yk

[Bryce, the CUDA Colonel @blelbach]

@orion_cole_2 @modal @nvidia @charles_irl I don't know yet exactly what I'll be working on as I haven't started yet.

What do you mean by automation stack?

[Bryce, the CUDA Colonel @blelbach]

@NickGideo @modal I think you were the only one so you get a prize, if you can find me in person at a future event!

[Bryce, the CUDA Colonel @blelbach]

I'm joining @modal!

Software is undergoing a revolution; we need new tools and infrastructure for this new era.

@modal is the platform I want in these times, so I decided to help build it.

If you want to come build it with me: https://t.co/UbkWFa7SzU and my DMs are open!

[ADSP: The Podcast @adspthepodcast]

📢 ADSP Episode 310 is out! 📢 In this episode, @blelbach and @code_report chat about Bryce leaving @nvidia, where he is headed next and more! 🥳 https://t.co/9geMXot8yk