Robert Sweeney

Today we’re launching Mercury 2 (https://lnkd.in/gP8pZuDA), the fastest reasoning LLM and the first reasoning dLLM: ~1,000 tokens/sec, ~5× faster than leading speed‑optimized LLMs. What I’m most excited about isn’t just the speed benchmark line, but what this unlocks for building real AI systems: multi‑step agents that don’t stall, voice assistants that can reason inside tight latency budgets, and long-form coding loops that stay in flow. Diffusion changes the generation loop: Mercury works more like an editor iterating in parallel than a typewriter committing one token at a time. We’re hiring engineers across research, systems, infrastructure, and product—If you’re excited about working on frontier LLMs and care deeply about engineering quality, DM me—I’d love to connect (https://lnkd.in/g28qHbNV).

149

8 Comments

If you're keen on real AI use cases, this is actually quite powerful and I encourage you to read further (and help us build: https://lnkd.in/gsYnAfKK) At Hex, we've found our Threads product (https://lnkd.in/gJKCurJB) is the best way to summarize customer support load, because we're not limited to what's in the customer support portal - we can ask in terms of active users, customer size, feature adoption, and still get great answers. This is based on a strategy of keeping your data warehouse as a source of truth, and all related data work in one place - Hex - so that AI understands concepts in your business and makes smart connections between things. Arguably, ALL data analysis for a particular vertical is eventually best achieved in a tool like Hex, connected to curated data in your data warehouse. Wait - can't I do this directly in the customer support tool? Can't I connect an MCP server and get a functionality like this in one of my other AI enabled products? Maybe, but you have 100+ SaaS apps, so play this scenario out: - How many places are you going to sync your core data sets? - How many cross product integrations are going to exist, and how many of them are you going to enable? (100 choose 2 is ~5000) - In how many of these products will all users have seats? - In how many of these products will these seats have permissions to use AI features? (Our support tool just told me I don't have permission to ask the agent questions, probably because I don't have the right seat type, which means more $$) My bet is a on a world with ONE place for all analysis.

33

2 Comments

This operationalizes something Andrej Karpathy prototyped late last year, the idea of an LLM "council." This is not just multi-model routing. It's multi-model reasoning. One query, multiple frontier models at the same time, then a synthesizer surfacing where they agree, where they differ, and why. The important signal here is the pattern underneath: model selection and cross-model reconciliation is moving from an expert workflow to a mainstream product capability. Feels like an early glimpse of how "AI decision support" will evolve. Less about picking the best model, more about orchestrating perspectives and resolving uncertainty. Worth watching where this pattern shows up next.

9

1 Comment

Waymo’s recent robotaxi incidents surface one of the most important and least discussed questions in AI product design: What should a system do when it doesn’t know what to do? Waymo’s answer is refreshingly honest, i.e. stop. Its vehicles are explicitly trained to halt when uncertainty rises (power outages, broken traffic lights, unexpected human behavior). This isn’t a model failure. It’s a product decision. When the safest action is inaction, the system chooses restraint over guesswork. At the level of one car, this is rational. At the level of ten cars, still sensible. At the level of hundreds, it becomes a system-wide failure. Waymo optimizes for local correctness, i.e. when unsure, do no harm. But without orchestration above the vehicle, local safety becomes global brittleness. Correct behavior in isolation can cascade into city-scale paralysis. The real lesson: building AI products isn’t about maximizing intelligence. It’s about designing systems that behave predictably when intelligence runs out and understanding how those behaviors compound at scale. Contrast this with China’s Apollo Go (Baidu) and Tesla and you will see three different strategies with different trade-offs. Baidu reduces ambiguity upstream with an “infrastructure-first” strategy. They’re not just building autonomous cars, they’re building V2X cities where roads, signals, and networks talk to vehicles. When things break, the city steps in. Fleets are centrally directed into predefined “safe harbor” states. The result is system-level resilience albeit at the cost of heavy coordination, regulation, and long-term capital alignment. Tesla takes the opposite bet. It assumes ambiguity will reach the vehicle and designs for it. No smart roads. Vision-only autonomy with end-to-end neural networks learning from millions of edge cases. Tesla isn’t optimizing for certainty in the moment, but for learning at scale over time. The product takeaway is clear: robust AI systems can’t rely on a single philosophy. The most resilient products layer their defenses: + Local safety behaviors that fail predictably + System-level coordination that prevents cascading failure + Continuous learning loops that improve with scale In AI, uncertainty isn’t the edge case, it’s the design constraint.

15

3 Comments

🔥 𝗙𝗶𝗿𝗲𝘀𝗶𝗱𝗲 𝗖𝗵𝗮𝘁 𝗼𝗻 “𝗕𝘂𝗶𝗹𝗱𝗶𝗻𝗴 𝘄𝗶𝘁𝗵 𝗔𝗜 𝗔𝗴𝗲𝗻𝘁𝘀” Had an incredible evening hosting a fireside chat with Matt Kraning (𝗙𝗼𝘂𝗻𝗱𝗲𝗿 & 𝗖𝗧𝗢 𝗼𝗳 𝗘𝘅𝗽𝗮𝗻𝘀𝗲, 𝗲𝘅 𝗖𝗧𝗢 𝗣𝗮𝗹𝗼 𝗔𝗹𝘁𝗼 𝗡𝗲𝘁𝘄𝗼𝗿𝗸𝘀) and a group of technology and product leaders from some of the world’s biggest 𝗰𝗼𝗻𝘀𝘂𝗺𝗲𝗿 𝗯𝗿𝗮𝗻𝗱𝘀, 𝗵𝗲𝗮𝗹𝘁𝗵𝗰𝗮𝗿𝗲 𝗰𝗼𝗺𝗽𝗮𝗻𝗶𝗲𝘀, 𝗮𝗻𝗱 𝗳𝗶𝗻𝗮𝗻𝗰𝗶𝗮𝗹 𝗶𝗻𝘀𝘁𝗶𝘁𝘂𝘁𝗶𝗼𝗻𝘀 in the Bay Area. We discussed one of the most urgent questions facing every product and engineering organization today: building 𝗔𝗜 𝗮𝗴𝗲𝗻𝘁𝘀 𝗳𝗼𝗿 𝘆𝗼𝘂𝗿 𝘂𝘀𝗲𝗿𝘀, and 𝗹𝗲𝘃𝗲𝗿𝗮𝗴𝗶𝗻𝗴 𝗔𝗜 𝗮𝗴𝗲𝗻𝘁𝘀 𝘁𝗼 𝘁𝗿𝗮𝗻𝘀𝗳𝗼𝗿𝗺 𝗵𝗼𝘄 𝘆𝗼𝘂𝗿 𝘁𝗲𝗮𝗺𝘀 𝗼𝗽𝗲𝗿𝗮𝘁𝗲, 𝘀𝗵𝗶𝗽, 𝗮𝗻𝗱 𝘀𝗰𝗮𝗹𝗲. 𝗔𝘁 𝘁𝗵𝗶𝘀 𝗽𝗼𝗶𝗻𝘁, 𝘁𝗵𝗲𝗿𝗲 𝗶𝘀 𝗻𝗼 𝗽𝗹𝗮𝘆𝗯𝗼𝗼𝗸. The companies that win won’t be the ones waiting for others to define it - they’ll be the ones learning, experimenting, evolving, and writing it themselves. Because the reality is: Your competition is already re-architecting their business around AI agents - driving higher efficiency, faster iteration, and reinvesting those gains to build better products. The next era of product leadership isn’t about adopting AI. It’s about adopting an 𝗔𝗜-𝗮𝗴𝗲𝗻𝘁 𝗺𝗶𝗻𝗱𝘀𝗲𝘁. Grateful to everyone who joined and contributed such sharp perspectives and open conversations. Deepak Kumar P Vache Moroyan DB Banerjee

186

Please check out Gemini Embeddings 2, our first natively multimodal embedding model. Mapping text, image, video, and audio into one unified space solves so many "plumbing" issues for AI developers. It’s awesome to see this evolve from a research goal to a production-ready tool. Can't wait to see what the community builds with it. Blog: https://lnkd.in/g4GKRe_N

197

2 Comments