Currently: @FuriosaAI, an AI chip startup Formerly: Meta, Yahoo, Veoh (don't ask), Mexico City Times (RIP), ABC News Once: @Cambridge_Uni, @PomonaCollege
Los Angeles
Joined May 2009
- Tweets759
- Following1K
- Followers642
- Likes7K
Pinned Tweet
I just learned an amazing, wonderful word: verschlimmbessern.
But I'm pretty sure I could make it a tad better... xitter.cf/Anne_Kreft/status/9946…
Oliver Libaw retweeted
🛬 Next week, we’re hosting our inaugural happy hour: Powered by Furiosa.
🍹 On August 8, we’re bringing together industry leaders and practitioners at the forefront of AI to discuss, share knowledge, and enjoy great food and drinks.
🤝 If you work in AI research, engineering, or infrastructure, come connect with the NYC community shaping AI compute, performance optimization, and system architecture.
RSVP right here: linkedin.com/events/73559606…
Oliver Libaw retweeted
We're proud to announce that we've raised $125M in Series C bridge funding to scale sustainable AI compute and accelerate our roadmap.
This brings our total funding to $246 million and will help us meet growing demand from global enterprise customers for our flagship AI chip, RNGD (pronounced “Renegade”).
“LG AI Research's adoption of RNGD has validated our chip’s performance and superior energy efficiency,” said Furiosa CEO June Paik. “Now we are scaling RNGD production globally to meet growing customer demand and applying learnings in the development of our next-generation chip.”
📰 Learn more: furiosa.ai/blog/announcing-f…
Amazing product. congrats @yutori_ai! Curious about the loss function used to minimize doomscrolling, though...
Oliver Libaw retweeted
Efficient AI inference isn't just an engineering detail — it's the key to unlocking the next generation of AI capabilities.
From agents and complex reasoning systems to enterprise-scale RAG deployments, the future of AI depends on powerful, cost-effective inference.
Learn why, and see how our Tensor Contraction Processor is leading the way: furiosa.ai/blog/furiosaai-fu…
#inference #datacenter #efficienctAI
sincerely quite pleased to be on one of @Scobleizer's AI lists... He's really got an amazing collection of people & accounts doing interesting stuff in AI.
Getting more and more of these in my Twitter DMs.
Always nice to be part of a thoughtful, informative exchange of ideas here on X.com...
In related news, I'm spending a bit more time on Bluesky these days (oliverlibaw.bsky.social)
This is a really cool achievement for @FuriosaAI: Our engineers' paper were accepted at #ISSCC 2025, a very competitive conference known as the "Chip Olympics."
It's a pleasure to work with such talented people building innovative AI inference hardware.
It has been a busy week for research paper announcements at FuriosaAI 📚✍️📖💪.
We're proud to announce that our paper, "RNGD: A 5nm Tensor Contraction Processor for Power-Efficient inference on Large Language Models", has been accepted as a Regular paper at #ISSCC 2025 - often called the “Chip Olympics” in our industry.
Being selected for the “Invited Industry” session at this highly competitive @IEEEorg conference is a significant recognition of our team's hard work and innovation.
We're looking forward to presenting our work to leaders in the field this February in San Francisco.
Learn more about RNGD (pronounced Renegade), which is in production now and sampling with customers: furiosa.ai/rngd.
Oliver Libaw retweeted
Startup FuriosaAI debuts RNGD chip for LLM and multimodal AI inference ift.tt/cv5VNmC
Oliver Libaw retweeted
🚀 Today at Hot Chips, we’re unveiling RNGD - data center AI accelerator for enterprise! furiosa.ai
💡150W TDP, Tensor Contraction Processor, and 48GB HBM3 make it powerful, efficient, and programmable in a single product - a GPUs have long struggled to achieve.
Oliver Libaw retweeted
We are honored to present at Hot Chips @hotchipsorg, an industry-leading event for high-performance computing innovations, on Aug 26 at @Stanford.
FuriosaAI co-founder and CEO June Paik will announce updates on tackling key challenges in LLM inference. hotchips.org/advance-program…
Oliver Libaw retweeted
Don't miss FuriosaAI CEO June Paik's presentation at AI Hardware Summit next month.
We'll be sharing exciting updates about our work to build inference solutions that deliver performance, programmability & power efficiency in one product. aihwedgesummit.com/events/ai…
@aiandsystems
Oliver Libaw retweeted
The short video of our ISCA presentation, “TCP: A Tensor Contraction Processor for AI Workloads” is available now. Our paper was selected as one of 4 papers out of 26 submissions from 18 companies this year for the ISCA industry track. youtube.com/watch?v=XKTWKCh9…
Hi! Here's an AI art project that doesn't plagiarize other people's drawings and paintings. I used an LLM to produce ~200 made-up descriptions of the most nondescript piece of art I could think of: a piece of duct tape stuck to a wall.
Foolishly breaking my self-imposed, much-ignored Twitter silence in order to share a silly thing I've built:
Behold the AI-powered Squirrel Squirter, custom object detection model running on a Raspberry Pi 4 wired to a motorized squirt gun. 🍊🐿️🔫🎯
#AI #opencv #raspberrypi
Lesson 4. When they say the Raspberry Pi is a resource-constrained device, it turns out they are correct. I trained a really accurate YOLO model on my Mac only to discover it crashed my Pi before it even loaded it completely.