Mark Lee

As AI workloads eat up global computing supply, DRAM prices are surging like never before—creating major challenges for teams trying to plan out their data center spend. One of my portfolio companies, Mext, is addressing the largest cost component in the datacenter: server memory (DRAM). From being on the board of Redis for the last twelve years, I’ve learned a lot about the various aspects of this problem and approaches to solving them. Mext found a breakthrough, using new AI techniques, for dramatically reducing the amount of server-DRAM required to run applications, all while maintaining performance. It intelligently manages the server’s memory, keeping hot pages (i.e., those in use by applications) in fast DRAM and offloading cold pages (i.e., those less used) to a much less expensive memory tier (e.g., NVMe Flash). Key to the approach is ensuring cold memory pages that are about to be accessed by an application are back in DRAM before the application needs them or notices that they were gone. This is done without modifications to the application or the OS – so it can run in the cloud or on-premise. Mext thus allows applications to either run using less DRAM or keep their DRAM footprint but do more with it. They're hosting a webinar along with Fred Weber (former CTO of AMD) on Jan 22nd at 10am pacific time. You can register here: https://lnkd.in/gXe9GCeE

153

3 Comments

Chatting with ZEDEDA CEO, Said Ouissal, and he has an elegant way of explaining the evolution of industrial Edge. Zededa provides OS for orchestrating Edge AI. 2010s were about IoT/Edge Connectivity. 2020s were about Edge computing/processing. 2030s will be all about Edge AI.

36

4 Comments

Last year, we invested in Cognichip: what we expect to be a generational semiconductor company by making chip design as simple and accessible as vibe coding. They came out of stealth today with a vision for how we will design chips in the future. Today, building chips is nowhere nearly as simple as uttering these commands. In fact, this very task requires going through thousands of pages of documents and hundreds of hours of experienced engineer time. It gets worse with every tech node. Chip design is tedious, which is why chips cost a fortune to design, and take forever to bring to market, resulting in only a handful of (giant) chip companies. Faraj Aalaei, who personally built, grew, and took several chip companies public over the past 3 decades, founded Cognichip with Simon Sabato and Ehsan Kamalinejad to disrupt this status quo: https://lnkd.in/gAkM_k6Q He sounded the alarm a decade ago, and as a dear friend of Lux, he was kind enough to share his vision that became Cognichip: Cognichip has attracted Olympiad-medalist mathematicians and veteran chip designers to make chip design as accessible as coding apps. We expect Cognichip to enable a future where there are as many chip companies as there are software companies, rapidly accelerating compute. We are excited to embark on this journey with the fabulous team Faraj has assembled, along with our friends at Mayfield and FPV Ventures

121

11 Comments

The magic of custom silicon happens at microscopic scale. Our AWS Trainium chips built for AI training and inference are architectural marvels, each capable of trillions of calculations per second with billions of transistors working in perfect harmony. Check out this interactive feature that explains how our chip designers think like city planners, optimizing every nanometer to move data at near light speed. https://lnkd.in/gCajddmg This attention to detail lets us deliver the performance and efficiency our customers need for their most demanding AI workloads. 

1,483

43 Comments

GPUs have gotten faster, and models have gotten larger. But storage is still stuck in the past. Modern AI workloads need instant access to huge datasets. Instead, teams waste time on slow retrieval, idle GPUs, and costly pipelines. Archil is changing this. It delivers object-storage scalability at block-storage speeds. The platform allows for seamless, high-performance access to massive datasets. By eliminating cold starts and accelerating throughput 30X, Archil enables teams to fully utilize their infrastructure and move faster on training, analytics, and AI deployment. Archil’s founder Hunter Leath brings rare technical depth and operating experience from building Amazon EFS and optimizing Netflix’s cloud performance. This is a team building the foundational layer that AI workloads desperately need. I’m thrilled to lead Archil’s seed at Felicis with Nancy Wang and support their mission to reinvent cloud storage for the AI era. https://lnkd.in/gWueXhzz

65

2 Comments

AI compounds improvements. In August we continued to deploy AI to build AI for investment alpha. August AI product releases from Decasonic include: 💪 Today, we stand at 125 AI teammates to 4 human teammates, a ratio of 31:1 🤖 We shipped +45 agents, 56% MoM growth, above plan ⚖️ We launched our AI Due Diligence Team, comprised of 35 agents performing human quality due diligence at 30x+ speed 📲 We integrated our 10 AI applications in an AI Operating System, unifying how they orchestrate within our workflows 📊 We added +10 agents in our Data Team, focused on on-chain, sentiment and technical market data 🌃 Kudos for the team's hard work, 9-9-7, nights and weekends on top of their day jobs, to ship world class AI. We now have AI that works 365-24-7 More to come...

38

4 Comments

Cerebras is now a generational AI company. But between 2018 and 2019, one tiny component nearly killed the company. Back then we were deep into building the first wafer-scale system. Everything depended on a single part most people will never notice. The interposer. It sits between the main board and the dinner-plate-sized silicon wafer. It has to carry enormous power and data loads while having to survive massive thermal cycling. Cold to hot. Off to fully lit. Over and over again. We had just built the largest chip in the world. There was no existing supply chain for it so we had to single source this part. That alone is terrifying. One vendor, no backup. If it failed, the system failed. If the system failed, the company could have failed. For almost three years, this thing was the boogeyman in every board meeting. Would it maintain electrical continuity? Would it warp under load? Would it survive burn-in? Would it work at scale? For long stretches, the honest answer was “we don’t know.” We were burning millions of dollars per month. Time was not our friend. A redesign could have pushed us far enough back that the market might have moved on. At one point, we pulled in a group that I first worked with at IDEO and that used to be called "Failure Analysis". That is not who you call when things are going well! We went back to first principles. Physics. Materials science. Coefficients of thermal expansion. Power dissipation per square millimeter. Analyzing what happens when you push this much power through a system at these temperatures. Again and again. Slowly, painfully, we got there. Finally the interposer worked. Every company pushing the frontier runs into problems like this. There are rarely quick, clever fixes. You just have to do real failure analysis, over and over, until the answer starts to reveal itself. The Cerebras team is itself built from the strongest, toughest material I’ve seen. They earned every inch of the other side of that problem.

335

26 Comments

Great speaking with Maria Bartiromo on Fox Business Network. We covered a lot of pressing semiconductor topics, including: - The importance of creating export policies that balance economic and security objectives. - How the U.S. Department of Commerce’s LOI investment of $150M in xLight Inc. will help drive chip innovation here at home. - Why new technologies, like extreme ultraviolet lithography, should allow us to wake up Moore’s Law. - And the critical role energy plays in advancing our AI capabilities. https://lnkd.in/gQtzJ9VB

127

8 Comments

Inertia is the commercial fusion energy company, and today we're taking a big step forward with $450M in funding! With support from Bessemer Venture Partners GV (Google Ventures), Modern Capital, Threshold Ventures, Neo, Uncork Capital, Long Journey, and WndrCo we're taking proven physics from the lab to build the industrial foundation for gigawatt-scale fusion power plants. The road to get here involved $30 billion in government funding for inertial confinement fusion experiments, the groundbreaking demonstrations of fusion that works at the National Ignition Facility at Lawrence Livermore National Laboratory, the incredible ingenuity of my co-founders Andrea (Annie) Kritcher and Mike Dunne, and the team of DO-ers we’re building at Inertia. Now, with our funding, we're going to: 🔆 Build and scale the production of the world's most powerful laser  🔆 Build the world's first fusion fuel factory to mass produce fusion targets 🔆 Design and build the first gigawatt, utility-scale fusion energy power plant We're bringing star power to Earth. Read the founders' letter: https://inertia.com/news/ Read the press release: https://lnkd.in/gVEEYEpv And come build with us: https://inertia.com/jobs

1,176

116 Comments

"Build the event you wish existed" That's what we at Monadical did last week at #NYTW. We wanted an AI Engineers discussion for Engineers. There's always a lot of events out there for VCs to network, or for startups to learn about G2M, but very little on lessons learned from actual engineers in the field. So that's the event we hosted. Big thank yous to our awesome panel, Roy Pereira, Ben Cohen and Corey J. Gallon. Here's the key takeaways and the AI tools we're using. 🚀 All three panelists independently called AI Agents the most transformative LLM application they’ve used. They specifically called out Claude 3.5 Sonnet for its accuracy and reliability. 🪨 We dug into how LLMs are “jagged”, not general. They can be shockingly good at some tasks and completely fail at others. Everyone agreed: good evaluations are critical (and hard.) 🧪 Corey noted how public benchmarks and reality are two different things. Most public evals are saturated or gamed. ♊ Ben emphasized that AI projects are actually two projects: building the tool and building the evaluation process. 🧱 We explored how falling dev costs may impact startup defensibility and labor demand. Roy shared that founders are already shifting strategies in response. ⚒️ In a world of daily AI launches, the panel discussed how they decide what’s worth attention, and what’s just noise. They called out tools like Goose, Aider, Claude Code, and Monadical’s own Cubbi, which helps run agentic workflows safely in dev environments. (links in the comments). CTA: What would you want to hear in an AI Applied Engineering talk you attended?

33

5 Comments

In the latest episode of the OpenAI podcast I got talk to OpenAI CFO Sarah Friar and legend himself Vinod Khosla about the AI industry. Sarah explained the relation between compute and revenue (OpenAI's revenue keeps 3x'ing each year) and Vinod provided an insightful way to tell if we're in a bubble: API calls. Are they going up or down? (Spoiler: Up, up, up!) https://lnkd.in/g2tTQBdp

22

1 Comment