S13 Bonus: Why Local On-Device AI Compute is the Future of Privacy with Behnam Bastani, Co-Founder & CEO of OpenInfer
Aug 26, 2026 · 27m
Summary
Benham Bastani, CEO of OpenInfer, discusses building an inference OS for the agentic AI era. He shares his journey from Meta and Roblox, emphasizing system-level thinking to optimize AI performance across diverse hardware. The episode covers pivoting to serve NeoClouds, the critical role of team harmony, and advice for entrepreneurs on clearly defining their problem and unique value.
Topics discussed
Intro: Cohesity ad and Code Story podcast theme
Sponsor ads: Tiger Data, Protected Harbor, Render
Benham Bastani's background and Meta VR work
Building Oculus Link and moving to Roblox
Vision for agentic AI and founding OpenInfer
MVP development and pivoting to NeoClouds
Strategic partnerships with Fortune 500 clients
Sponsor ads: Tiger Data and Cohesity
Product roadmap and sovereign cloud trends
Scaling strategy and the Tesla Roadster analogy
Team building, culture, and hiring philosophy
Sponsor ad: Render
Approaches to scaling and managing ambiguity
Proudest achievements: Team, architecture, vision
Mistake: Targeting premature robotics market
Future of AI inference and vertical applications
Leadership influences and advice for entrepreneurs
Outro and sponsor ads: Dropbox, Capital One
Listen ad-free on Castria