Jon Gelsey

Jensen Huang said something at #GTC2026 that I haven't been able to stop thinking about. 🦞 "Every company in the world today needs to have an OpenClaw strategy, an agentic system strategy. This is the new computer." Yes, the NemoClaw launch is very cool and I'll be trying it out as soon as possible (obviously). But what I think is even more interesting is what is really behind all of this. His keynote, the GitHub repos, social media hype, the billion-dollar infrastructure bets - We are living through a fundamental shift in how work gets done. We are moving from tools to agents. Can you feel it? A tool does what you tell it but an agent can decide what to do next. And after that. And after that. If that sounds like a small upgrade to you, you're thinking about it wrong, and IMO most organizations are still thinking about it wrong. Andrej Karpathy published autoresearch on GitHub earlier this month. One agent, one metric, one instruction: never stop, never ask for permission. It runs experiments overnight while you sleep and you wake up with 100 completed iterations. So what does that mean for us? The human's job shifts from doing the search to defining what's worth searching for. Jensen made another prediction. He said every engineer will have an annual token budget worth roughly half their base salary - so they can be amplified ten times. He called tokens the new recruiting currency in Silicon Valley. Not equity. Not perks. Compute. Access. Think about what that means. Compensation is starting to reflect not just what you know or how long you work, but HOW MUCH you can do with the intelligence available to you. That's honestly an incredible shift. Most leaders I talk to know something big is happening. Fewer have sat with the hard questions: How do we measure output when the process runs itself? Who is accountable for an agent’s decision? How do we manage a workforce where half the "employees" are digital agents? Those really aren't technology questions - they are leadership questions. And I haven't seen a lot posted about this yet. (And no - AI did not write this. I did.) Where is your organization on this? How have you seen companies building their agentic strategy and what does it look like? Drop your thoughts in the comments. https://lnkd.in/eEY_pWxs

54

33 Comments

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

AI infrastructure has hit a power wall — and interconnect inefficiency is at the core. As bandwidth and connectivity requirements explode, copper is no longer keeping pace. Co-packaged optics changes that. Today, I’m proud to share that Ayar Labs has raised $500M in Series E funding to accelerate volume production of co-packaged optics for AI scale-up. I want to sincerely thank our lead investor, Neuberger Berman, and our new and existing investors for their conviction and partnership — including ARK Investment Management LLC, Insight Partners, Qatar Investment Authority, Sequoia Capital Global Equities, 1789 Capital, and our strategic partners AMD, Alchip Technologies, MediaTek, and NVIDIA. This funding fuels the next phase of our execution to: - Scale high-volume production and test capacity - Expand global operations, including our new office in Hsinchu, Taiwan - Deepen ecosystem partnerships across the AI and custom ASIC landscape - Accelerate the deployment of our CPO solution Thank you to all of our investors, customers, and most importantly, the Ayar Labs employees — your vision and execution make breakthroughs like this possible.

89

17 Comments

Quantinuum’s planned IPO is the moment quantum finally hits the public markets in grown‑up form. This isn’t another early‑stage SPAC story; it’s a full‑stack platform spun out of Honeywell, with trapped‑ion hardware that regularly tops industry benchmarks and a real software and applications business on top. By combining industrial‑grade engineering with production‑grade tooling in chemistry, security, and optimization, Quantinuum gives investors exposure not just to “qubits,” but to an integrated quantum stack that already serves enterprise workloads. For technical and business leaders, as well as investors, the key question now is how public markets will price a quantum company that actually looks like an infrastructure and software provider rather than a speculative science project. Whatever the final valuation, this IPO will likely become the benchmark for what “mature” looks like in quantum—deep IP, differentiated hardware, and diversified revenue experiments before ringing the bell. Headline: Honeywell’s Quantinuum Quantum-Computing Unit Plans IPO – https://lnkd.in/euZjkGtG

194

5 Comments

Thank Dani Cherkassky Great piece on the architecture shift voice AI needs. The core thesis: cloud-centric voice systems fail in real-world conditions because they are too slow, too expensive to keep always-on, and deaf to spatial context. The solution is a hybrid architecture—fast, always-on edge processing (Spatial Hearing AI and Cognition AI) handling 80% of interactions locally, with cloud LLMs reserved for complex reasoning. But the Kahneman thesis runs deeper than the article suggests. System 1 and System 2 are not just useful metaphors—they reflect an actual biological sequence. Fast, intuitive processing (System 1) evolved first; slow, deliberate reasoning (System 2) was layered on top. The brain's architecture reflects this: spatial-acoustic processing happens in milliseconds at the brainstem and primary auditory cortex, long before language centers engage. Kardome is not just borrowing Kahneman as marketing. Their architecture is neurologically correct—it replicates the hierarchical, parallel, spatially-aware processing that makes human auditory cognition so robust. Here is the deeper point: hearing preceded language in human evolution, and spatial hearing preceded both. The mammalian auditory system evolved to detect predators and locate prey—survival functions that required sub-second latency and continuous environmental awareness. Language processing was layered on top of this infrastructure. Current voice AI inverts this evolutionary logic: it starts with language models and treats spatial audio as an afterthought. Kardome restores the natural hierarchy. 🙌 Danny Shapiro, Rory Sutherland

35

2 Comments