Michael Stewart
@mps_vcScientist-engineer turned VC at @M12vc: Generative AI, gaming, semiconductors, materials science, and all things deep tech. h-index: 31
Similar User
@_TheResidency
@adriannewman21
@AnnChien
@bestape
@keroserene
@CarmillaEtc
@yosriady
@nicksusi
@agartha_one
@nick_linck
@sirarthurmoney
@mald0xx
@regenagvocate
@GiacobettiMegan
@Jesse__McKee
"Coscientist"—a GPT-4 based autonomous LLM system that demonstrates appreciable reasoning capabilities, ... solving of multiple problems and generation of code for experimental design" nature.com/articles/s4158… @daniil_boiko @gabepgomes @CarnegieMellon
Katsumi Ishizuka AD & SB (Animation Direction & Storyboard) Vincent Chansard & Yen Bm/Chris Genga
👀 instacart like exits result in different outcomes for VCs at early vs growth stage @BarisGSF
Venture capital thrives on seeing successful exits. When the IPO pipeline clogs, it's not just founders and investors who suffer. It's the entire tech ecosystem. ServiceTitan is one of the first VC-backed SaaS companies going public for a long period of time, hence making it a…
AI is the new input for every company to expand their output, and the rapid adoption of this technology and the intensity of its everyday use are the most important ways companies can get ahead today. Here are highlights from my keynote at #MSIgnite.
We’re honored to have been recognized by @satyanadella during his keynote at Microsoft Ignite! Microsoft has been an incredible partner, and we’re especially grateful for the experience and vision that Pradeep Nair, our Founding CTO, brings to Armada after over a decade at…
The answer to this riddle is also evals
Peter Thiel: we have 100 times as many PhDs as we did in 1924 but scientific progress has slowed or stayed the same, so a PhD today is 99% less productive
“The first of it’s kind Corsair compute platform brings blazing fast token generation for high interactivity applications, with an emphasis on making Gen AI commercially viable.” - d-Matrix tinyurl.com/n3z4bmsw
Delighted to see @M12vc company @armada_ai highlighted in the infrastructure section of @microsoft #Ignite2024 keynote! Next generation solutions for edge AI, extending Azure to endpoints worldwide via @Starlink
.@MSFTResearch and collaborators are proud to present BiomedParse, a practical and true foundation model for image analysis across 9 different medical imaging modalities. In @naturemethods nature.com/articles/s4159…
Hazard ratio=0.07❗️ That's the ridiculous effect of tirzepatide (dual GIP/GLP1r agonist) on the 3-year incidence of diabetes among patients with obesity It's the kind of massive effect you expect for vaccines, not chronic disease prevention. nejm.org/doi/full/10.10…
Moondream raises $4.5M to prove that smaller AI models can still pack a punch venturebeat.com/ai/moondream-r…
Happy ‘screaming partially muscled skeleton’ day to all who celebrate.
Old growth forests of the internets: "putting all this together, searching IMG_XXXX offers the most authentic social feed ever seen on the Internet- in video, no less!" ben-mini.github.io/2024/img-0416
The ChatGPT desktop app for Windows is now available for all users 🖥️ Get faster access to ChatGPT with the Alt + Space shortcut, and use Advanced Voice Mode to chat with your computer and get hands-free answers while you work. openai.com/chatgpt/desktop
An article claiming Ilya sutskever said: Results from scaling up pre-training — the training phase where an AI model learns from vast unlabeled data — have plateaued. Here is stronger evidence of what Ilya thinks of scaling LLM.
100% The secrecy and obscurity is helping no one. More Bell Labs would be great, where do the innovations go? If we stimulate that much new invention, which we should, the "lab to fab pipeline" must become more of a conveyor belt. As much as I love R&D and funding more semis,…
We need multiple bell labs now to accelerate post-moore's law technology invention and deployment. Current silicon valley hardware startup culture doesn't encourage openness and sharing. We have 25+ AI hardware startups competing with each other, redoing the same tech stack…
At @M12vc we're proud to be invested in George and Ash and the @fastinoAI team. Eliminating hallucination, keeping response latency below the reflex threshold, shifting unnecessary GPU overhead to CPUs, and creating models tailored to tasks is the way forward.
Today we are emerging from stealth to launch a new family of LLMs with a novel architecture. Unlike traditional LLMs, Fastino’s task-optimized models are engineered to deliver consistent, accurate output with near-instant inference on CPU and flexible deployment options. (1/4)
United States Trends
- 1. Dodgers 51,2 B posts
- 2. Lakers 38,2 B posts
- 3. #DWTS 75,3 B posts
- 4. Chandler 39,7 B posts
- 5. Dylan Harper 1.621 posts
- 6. #kaicenat 5.007 posts
- 7. Duke 37,9 B posts
- 8. Ilona 13 B posts
- 9. Joey 27,3 B posts
- 10. Sasaki 8.885 posts
- 11. Matt Allocco N/A
- 12. Ohtani 7.168 posts
- 13. Soto 25,7 B posts
- 14. #IslamabadMassacre 418 B posts
- 15. Cooper Flagg 5.101 posts
- 16. Suns 19 B posts
- 17. Beal 7.279 posts
- 18. Glasnow 3.040 posts
- 19. Kershaw 1.733 posts
- 20. Buehler 1.332 posts
Who to follow
-
The Residency
@_TheResidency -
Adrian
@adriannewman21 -
Ann Chien
@AnnChien -
Kyle | bestape - ⚖️/acc @🌉 🖖
@bestape -
serene ❄️🎹
@keroserene -
Carmilla
@CarmillaEtc -
Yos 💫
@yosriady -
Ņ̨͈̯̻̪̥̥ͅick S̘͓͙usi
@nicksusi -
Grid Free Minds
@agartha_one -
Nick Linck
@nick_linck -
lz.sirΔrthurmoney()
@sirarthurmoney -
MalD0xx
@mald0xx -
🍉🍉🍉
@regenagvocate -
Megan Giacobetti
@GiacobettiMegan -
Jesse.R.McKee
@Jesse__McKee
Something went wrong.
Something went wrong.