Security may become the lion's share of global compute spend, and Socket is emerging as one of the most important security companies in the world
Really excited to partner with the General Compute team! Through working with companies like Decart, I saw how impactful unlocking new types of compute could be. General Compute focuses on making ultra fast compute and inference available for the best businesses in the world
@FPuklowski and @GoodisonJason are forces of nature, and consistently prioritize their customers and partners. I wouldn't bet against them
🚨 Big news today: We’ve raised $100M for our Series A! 🎉 🚀
We believe demand for fast inference is many times larger than demand for slow inference, and it’s about to go exponential.
Thanks to our co-leads @BlackbirdVC and @Abstractvc. Two visionary partners join our journey 🎉 🚀
Surprised Goldman won left lead on Anthropic given so much of the Ant finance team is from Morgan Stanley
Not sure who at @baseten spearheads their YouTube, but it’s just plain awesome
1) increases aura
2) highlights the technical depth of different team members
3) unique yet fun formats like HORSE and whiteboarding
Tesla is one incredible European designer away from owning the car market
Pay $25m for whoever did that Genesis rebrand, and it’s game over
David Kwon retweeted
Our heterogeneous cloud just got better. We're announcing a cluster scale deployment of @positron_ai as our next partner.
We believe demand for fast inference will outstrip supply for the next decade. Our bet on Positron is another step towards getting fast inference for opensource models into builders' hands.
Hardware goes live Q1 2027.
Thanks to @mitesh711, @trsohmers and the whole Positron team! We're excited for this partnership.
LFG!
it is possible to retweet someone's thoughts, like I'm doing here
VC investors have enabled the AI industry to bastardize revenue reporting.
In the 2010s, SaaS companies popularized reporting the metric of Annual Run Rate because it was a reliable measure of long term, contractual revenues into the future. The key is that the revenues are contracted and guaranteed thru the contract life hence why forecasting it forward is appropriate.
VCs have taken this to mean Annualized Run Rate when AI is concerned and where you take transactional/API revenue from the last month (or even last week in some cases) and forecast it forward.
It’s insane that investors have enabled this to happen but it’s happened.
Separately, revenue reporting in this whole industry is very insincere, and we should do away with revenue multiples and replace them with gross profit multiples
Valuing a data business' annualized GMV with a 25% take rate or an inference co's annualized revenue with 20% gross margins in the same way you would 80% software margin ARR is just full cope
Separately, revenue reporting in this whole industry is very insincere, and we should do away with revenue multiples and replace them with gross profit multiples
Valuing a data business' annualized GMV with a 25% take rate or an inference co's annualized revenue with 20% gross margins in the same way you would 80% software margin ARR is just full cope
With neoclouds rerating down, the revenue multiples for some of these inference businesses seem really silly
Thrilled to co-lead Volantis’ $88m Series A with @lachygroom.
Volantis has assembled an insane team to tackle the memory bottleneck: a major, persistent issue in AI. By using optics to target orders-of-magnitude gains in memory bandwidth and capacity, they open the door to much larger models and radically faster, cheaper agents.
Tapa (@semiDL) is an n-of-1 founder, and he obsesses over hardware and accelerating abundance. If you’re interested in learning more/joining, reach out!
Excited to announce Volantis's $88M Series A.
We are solving Al's memory bottleneck by using optics, enabling chips with huge amounts of fast & cheap memory.
By boosting both the memory bandwidth and capacity per chip by orders of magnitude, we enable ultra-fast inference (up to 10,000 tps/user) for large models (>10T) - with low $/tok to boot.
Initially, this will enable insanely fast agents - think coding agents that finish in minutes or even seconds instead of hours.
More excitingly, optics is a fundamentally scalable way to increase memory systems. Not 2X/year, but by orders of magnitude across new generations. This will enable a structurally new Al industry, including restarting scaling laws, holding entire repos in context windows & more.
Our team has pioneered many core semiconductor technologies: the 1st CoWoS product, early HBM, the 1st silicon photonics CPO systems, the 1st high volume tunable VCSELs, the 1st processors to directly communicate using light & more.
We’ve already sent data >10× farther than equally tiny electrical wires inside a chip package. Our next iteration is already taped out and targets world-record bandwidth density over relevant distances, read more: volantissemi.ai/news-insight…
David Kwon retweeted
Socket now protects four of the Magnificent Seven, two of the three hyperscalers, nearly every leading AI lab, and one of America's largest automakers.
This summer, @SocketSecurity automatically blocked two live supply chain attacks at the world's largest company.
AI coding agents are writing code faster than ever and pulling in more open source dependencies than ever. Socket is the guardrail that lets developers and agents move fast without shipping malicious or high-risk components.
Proud that Socket is protecting America's software supply chain, whether the code is written by humans or AI.
David Kwon retweeted
🚨 BREAKING: Super excited to announce we're deploying the world's fastest inference with @Cerebras.
Talk to any developer and they're excited to build with 20x faster AI... the problem is there's almost no compute available.
We're here to solve that.
Using GPUs for prefill - it's now more affordable than ever too.
Thank you to the whole Cerebras team and excited to grow this partnership.
First tokens live Q127 🚀