Episode

Interview #73 Jay Dawani, CEO of Lemurian Labs

Podcast
The Artificial Intelligence Podcast
Published
Aug 9, 2025
Duration seconds
1532
Processing state
not_requested
Canonical source
https://podcasters.spotify.com/pod/show/tonyphoang/episodes/Interview-73-Jay-Dawani--CEO-of-Lemurian-Labs-e36laoj
Audio
https://traffic.megaphone.fm/APO4882937706.mp3
JSON
/v1/public/podcasts/the-artificial-intelligence-podcast-3635848/episodes/interview-73-jay-dawani-ceo-of-lemurian-labs
Markdown
/podcast/the-artificial-intelligence-podcast-3635848/interview-73-jay-dawani-ceo-of-lemurian-labs.md

Actions

  • POST https://stenobird.com/v1/public/podcasts/the-artificial-intelligence-podcast-3635848/episodes/interview-73-jay-dawani-ceo-of-lemurian-labs/transcription-requests
    Idempotently request low-priority transcript generation for this episode.
  • GET https://stenobird.com/podcast/the-artificial-intelligence-podcast-3635848/interview-73-jay-dawani-ceo-of-lemurian-labs.md
    Read the agent-friendly Markdown representation of this episode resource.

Summary

Join Jay Dawani, CEO of Lemurian Labs, as he discusses the critical infrastructure challenges facing AI development and his company's efforts to rebuild the AI software stack from the ground up. Drawing from his experience as a former NASA AI advisor working on Mars Rover navigation and exoplanet research, Dawani explains how current AI systems are plagued by massive inefficiencies, with some data centers operating at only 10-15% utilization despite consuming enormous amounts of energy. The conversation explores how the industry must shift from kernel-based programming models designed for single GPUs to dynamic runtime systems that can efficiently manage communication and memory across hundreds of thousands of processors, ultimately making advanced AI more accessible and sustainable.