Saliya Ekanayake retweeted
800+ tokens/sec per user. Heterogeneous inference in action.
At #OCP2026, d-Matrix demonstrated live attention/FFN disaggregation across NVIDIA H100 GPUs and memory-centric Corsair™ nodes.
Watch the demo:
nextgeninfra.io/2026-ocp-glo…
Saliya Ekanayake retweeted
Tonight in SF: d-Matrix’s Nithesh Kurella and Prashant Nair join the 90/30 Club to discuss Raptor—the first 3D-DRAM accelerator for generative inference and what we’re learning from early silicon.
Up to 100 TB/s memory bandwidth per card.
Join us: luma.com/kyovs0is
#AI #AIInference #dMatrix
Saliya Ekanayake retweeted
AI Infra Summit went beyond the show floor.
We welcomed a group of reporters and industry analysts to the d-Matrix office and lab for a closer look at what we’re building, from our silicon to our server racks.
Thank you to everyone who joined us for the lab tour.
#InfiniteInference #AIInfraSummit
Saliya Ekanayake retweeted
Heterogeneous inference is here.
Stop by Booth 722 at #AIInfraSummit2026 to see how GPUs + d-Matrix are coming together to power AI inference at scale.
#AIInference #AIInfrastructure
Saliya Ekanayake retweeted
1/ Today, we announced Gimlet Labs’ $300M Series B, led by @a16z, joined by @SapphireVC as a major investor bringing our valuation to $3B.
We started Gimlet with a simple conviction: inference would become the dominant AI workload, and its infrastructure would need to be rebuilt from the ground up.
Saliya Ekanayake retweeted
Some conversations make it obvious we’re in a once-in-a-generation shift in AI.
@sidsheth shared his take on the watershed moments that brought us into the #AgeOfInference — and why breaking the memory & bandwidth wall now defines all future AI infrastructure.
Great to dig into it with Gavin Baker & Mark Wade… and yes, we all agreed @elonmusk’s idea of data centers in space actually makes a lot of sense. 😉
#AI #GenerativeAI #dMatrix
Saliya Ekanayake retweeted
When I created Claude Code as a side project back in September 2024, I had no idea it would grow to be what it is today. It is humbling to see how Claude Code has become a core dev tool for so many engineers, how enthusiastic the community is, and how people are using it for all sorts of things from coding, to devops, to research, to non-technical use cases. This technology is alien and magical, and it makes it so much easier for people to build and create. Increasingly, code is no longer the bottleneck.
A year ago, Claude struggled to generate bash commands without escaping issues. It worked for seconds or minutes at a time. We saw early signs that it may become broadly useful for coding one day.
Fast forward to today. In the last thirty days, I landed 259 PRs -- 497 commits, 40k lines added, 38k lines removed. Every single line was written by Claude Code + Opus 4.5. Claude consistently runs for minutes, hours, and days at a time (using Stop hooks). Software engineering is changing, and we are entering a new period in coding history. And we're still just getting started..
Lyrics I had written sometime back comes to life with Suno.ai.
Enjoy!
Pata Aran Akase | පාට අරන් ආකාසේ .... youtu.be/w1K-FEhcIW0?si=Ep4B… via @YouTube
#GrokImagine is absolutely mind-blowing for both the quality and speed of generation. Almost scrollable speed
Saliya Ekanayake retweeted
New paper: turns out you can train deep nets without normalization layers by replacing them with a parameterized tanh()
Growing up in Sri Lanka coffee was not a regular drink, but after being in Seattle for a while, and trying out freshly roasted beans and brewing at home, I just don't want to miss it ... psst mine tastes better than starbucks :)
Saliya Ekanayake retweeted
Easily the most impressive uncut autonomous video I've ever seen, from @physical_int
I created a video to help explain how LLMs work. Hope you'll find it useful!
youtube.com/watch?v=6o0Xnl6f…
P.S. Creating videos is hard, but kind of fun experience
Saliya Ekanayake retweeted
Hi all prospective grad students! Our Equal Access to Application Assistance (EAAA) program for
@Berkeley_EECS is now accepting applications! Any PhD applicant to @Berkeley_EECS can submit their application for feedback by Oct. 6th 11:59PM PST
forms.gle/rYVtXyaE9qE8ugST7
This seems like a cool research place to join mlcollective.org/?s=09
Saliya Ekanayake retweeted
Real2Code -- translating real-world articulated objects to sim using code generation! With the code representation, this method scales well wrt the number of object parts, check out the 10-drawer table it reconstructed 😉
Here’s something you didn’t know LLMs can do – reconstruct articulated objects!
Introducing Real2Code – our new real2sim approach that scalably reconstructs complex, multi-part articulated objects.
arxiv.org/abs/2406.08474























