Chief Scientist

Alexy Khrabrov
  • 5.0 (1)
  • Technology
  • Updated Daily

Tales of code, data, AI and open-source community from the heart of the Silicon Valley.

Episodes

  1. May 25

    Chief Scientist: Slava Tykhonov, CODATA Paris, May 2026

    Slava Tikhonov, Head of AI at CODATA, joins Alexy Khrabrov in Paris to discuss the semantic layer needed for reliable AI. Video Blog The conversation explores how standards such as Croissant, CDIF, DIDs, verifiable credentials, and ODRL can help AI systems find trusted data, understand provenance, enforce policies, and reduce non-determinism. Tikhonov explains how signed question-answer pairs, knowledge graphs, temporal versioning, and multilingual vocabularies can support reproducible AI for science, open compute, and enterprise deployments. Topics: CODATA, reliable AI, semantic layer, data provenance, Croissant, CDIF, DIDs, ODRL, verifiable credentials, knowledge graphs, AI agents, open compute, enterprise AI governance.

  2. May 24

    Chief Scientist: Sir Timothy Gowers, GOSIM Paris 2026

    Sir Timothy Gowers joins Alexy Khrabrov at GOSIM AI Paris to discuss mathematics, AI, and the future of proof. Video Blog Gowers explains why his Cambridge group studies automatic theorem proving not only as prompt engineering, but as a way to understand how humans solve mathematical problems. The conversation covers humane mathematical explanation, Lean and proof assistants, auto-formalization, AI systems checking journal proofs, the role of human judgment, and why people who deeply understand mathematics may have an even greater advantage as AI tools improve. Topics: Timothy Gowers, AI and mathematics, automatic theorem proving, Lean, formal methods, proof assistants, auto-formalization, mathematical understanding, AI education, GOSIM AI Paris.

  3. May 24

    Chief Scientist: Yann LeCun on a new Global AI Model, Paris 2026

    Yann LeCun discusses Tapestry, an open source AI initiative aimed at giving countries, regions, institutions, and communities a way to collaboratively train frontier models without pooling private data. Video Blog In this conversation with Alexy Khrabrov in Paris, LeCun explains why AI sovereignty will require open source collaboration, federated training, diverse local data, and broad participation from governments, universities, companies, and individual contributors. He also clarifies his position on LLMs: they are clearly useful for coding, math, and information access, but he does not see them as a path to human-level intelligence. Topics: Tapestry, open source AI, AI sovereignty, federated training, distributed model training, AI Alliance, LLMs, cultural diversity, democracy, frontier models.

  4. May 24

    Learning Rust from Custom-Made Books

    The new way of learning is in-domain, working with AI to both build a codebase you care about and the books that describe it and teach you the technology AI has co-authored. In this episode we build a meetup graph scraper in Rust and load it into three modern multimodal agentic databases, Falkor, Helix and Surreal. We ask Codex and Claude to write books about them, illuminating various Rust features and design decisions. Then we read the beautifully made PDFs in redwood groves, on mountaintops, and by the beach!

  • 4 Episodes

About

Tales of code, data, AI and open-source community from the heart of the Silicon Valley.

Information

  • Creator
    Alexy Khrabrov
  • Years Active
    2026
  • Episodes
    4
  • Rating
    Clean
  • Copyright
    © Alexy Khrabrov
  • Show Website
    Chief Scientist