Andrei Lyskov

Document-centric tasks sit at the core of many enterprise, business, and government workflows. Search is important, but the real challenge is going beyond retrieval, enabling systems to reason across documents, verify facts, and handle multi-step information tasks. Great to see a new model from Databricks moving in this direction. I expect we’ll see many more models, including open-weight ones, designed specifically for document-driven workflows, an area with huge potential across enterprise and government use cases.

23

1 Comment

This is why I love working at Databricks. Innovation here starts from first principles (and it’s never easy). Our product and R&D teams build “Useful AI” capabilities that enable organizations to generate real alpha, not just add to agentic hype. Add KARL to the list!

40

Why does the second search feel instant when the first one takes 3 seconds? I built a local semantic search engine using EmbeddingGemma. The bottleneck? Generating embeddings—10-100ms per chunk of text. Twenty chunks takes under a second. Two hundred chunks? A few seconds. Two thousand? You're waiting. And every search regenerated every embedding. Wasteful. The fix is simple caching: save embeddings to a JSON file after the first indexing run. On subsequent searches, load from disk instead of regenerating. First run: generate and cache. Every run after: instant. The real insight: caching doesn't just make it faster—it changes how you use the tool. Without caching, you think twice before searching. With it, searching becomes nearly free. That's the difference between a tool you built and a tool you actually use. I teach by building in public and sharing what I learn. Full write-up with code:  https://lnkd.in/gJtgEQ7s  #AI #MachineLearning #BuildInPublic #AILiteracy

4

Weekly Quack 011 -- the week's AI news in 246 words and ~96 seconds of reading time 👇🏼 xAI open-sourced Grok-2.5 on Hugging Face with Grok-3 coming in six months, though critics noted restrictive licensing. Amazon's AGI Lab defended its "reverse acquihire" strategy, Apple is discussing custom Gemini integration for Siri by 2026, Microsoft's Suleyman warned against AI consciousness research citing "AI psychosis" risks, and Meta announced a $100M solar partnership for its South Carolina data center. Silicon Valley launched well-funded PACs to influence 2026 midterms and counter regulation calls. Subscribe here 👉🏻 https://lnkd.in/g5GmFqi6 Headlines 🤗 Elon Musk's xAI open-sourced Grok-2.5 on Hugging Face, with Grok-3 planned for six months out. While increasing transparency, the AI community flagged "anti-competitive" license terms limiting usage. 🦾 Amazon AGI Lab head David Luan defended the company's "reverse acquihire" strategy of hiring startup talent and licensing technology rather than full acquisitions, calling it "perfectly rational" for accessing necessary resources. ♊︎ Apple is reportedly in talks to integrate Google's Gemini AI into future Siri versions. Discussions center on a custom Gemini running on Apple servers for privacy, potentially launching by 2026. 😵‍💫 Microsoft AI chief Mustafa Suleyman warned against studying AI consciousness, citing "AI psychosis" where users form unhealthy attachments to chatbots. He argues this could lead to misguided "AI rights" advocacy. ☀️ Meta will partner with Silicon Ranch on a $100M, 100-megawatt solar farm in South Carolina to power its first state AI data center, both coming online in 2027. 💰 Silicon Valley launched new PACs with $100M+ funding to influence 2026 midterms, supporting "pro-innovation" candidates and opposing strict AI regulation advocates.

16

1 Comment

“The AI-powered system is anti-fragile. Every user interaction, especially corrections, becomes a signal that enriches the system for everyone. This model enables continuous improvement. Traditional semantic layers are most accurate the day they're built, then slowly decay until someone rebuilds them.” In the very near future hopefully we’ll stop talking about these traditional semantic layers altogether. MotherDuck has come to the same conclusion as us at Fabi.ai : traditional semantic layers are a story told by vendors and they’re expensive, fragile and not all that useful in an AI world. True story: I was chatting with the head of engineering at one of the big BI firms peddling semantic layers, and he told me “to be honest, we see very little adoption, we’re mostly hoping people adopt it for AI” (so basically hoping AI needs it so they can sell their product)

7

7 Comments