Subvolts take Inference infrastructure is the new frontier, with custom models driving demand. May 1, 2026 - Developers
- Learn about the challenges and solutions for deploying and scaling AI models efficiently.
- Business
- Understand how specialized models and inference infrastructure create business value and competitive advantages.
A deep dive with a CEO of a fast-growing AI infrastructure company provides valuable insights into current industry challenges.
Baseten CEO Tuhin Srivastava on the AI Inference Crunch, Custom Models, and Building the Inference Cloud
Audio streams from Conviction. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
Baseten CEO Tuhin Srivastava joins No Priors hosts Sarah Guo and Elad Gil to discuss the AI inference crunch and the rise of custom models. Srivastava highlights Baseten's 30x growth and argues the application layer will persist. The conversation covers GPU constraints, multi-cloud strategies, long-term contracts, and the importance of the software layer for AI services.
AI models need powerful computers to run, and demand is booming. This episode explains how companies like Baseten build the systems to handle this demand, especially for specialized AI tasks. It matters because efficient AI use depends on this underlying computer power.
This episode offers insight into the critical infrastructure challenges facing AI development, specifically inference. It's for founders, engineers, and investors navigating the current AI landscape. By discussing custom models and multi-cloud solutions, it connects to how frontier labs like Google Gemini and OpenAI are pushing model capabilities while infrastructure providers grapple with scaling demand.
- HostSarah Guo
- HostElad Gil
- GuestTuhin SrivastavaCEO, Baseten
- Guest: Tuhin Srivastava, CEO of Baseten
- Show: No Priors
- Length: 43 minutes
- Baseten Growth: 30x
- Baseten Cloud Fabric: 18 clouds and 90 clusters
Baseten CEO and co-founder Tuhin Srivastava sits down with Sarah Guo and Elad Gil to discuss the rapid growth of AI inference demand, Baseten’s 30x growth, and why inference is becoming the strategic “last market.” Tuhin Srivastava argues the application layer will persist because companies with unique user signals can encode value into workflows and post-train specialized models, citing examples like Abridge and support workflows. The conversation covers GPU capacity constraints, Baseten’s multi-cloud fabric across 18 clouds and 90 clusters, long-term contracting dynamics, the importance of the software layer for stickiness, evolving workloads, multichip possibilities, and operational lessons at scale.
Sign up for new podcasts every week. Email feedback to show@no-priors.com
Follow us on Twitter: @NoPriorsPod | @Saranormous | @EladGil | @Tuhinone
Chapters:
00:31 Baseten growth
01:55 Why the app layer wins
05:57 Serving frontier customers
07:55 Open source model mix
09:21 Chinese models and geopolitics
13:07 Custom inference dominates
14:22 Post training acquisition
17:10 When to invest in custom models
18:35 Supply crunch and data centerse
22:25 Longer GPU Contracts
24:09 What Makes a Winner
26:07 Multi Chip Future
28:19 Runtime Roadmap
31:08 Scaling Edge Cases
33:48 Hiring and Leadership
36:44 Operations Pager Culture
38:19 Efficiency Drives Demand
40:41 Concierge Everything Future
42:34 Conclusion
- Inference
- Tuhin Srivastava
- The process of using a trained AI model to make predictions or generate outputs.
- Multi-cloud fabric
- Tuhin Srivastava
- A system that connects and manages resources across many different cloud computing providers.
39 min1 h 1136 minSame guests, other shows
59 minThe CEO Behind the Fastest-Growing AI Inference Company | Tuhin Srivastava
WithTuhin Srivastava
53 min
1 h 35Demis Hassabis on AGI, Robots Scale Production, and Elon’s $1T Mars-Shot Comp | EP #253
WithDemis Hassabis, Steven Kotler, Salim Ismail, Dave Blundin
1 h 02AI Agents: Mirage Or Real Revolution? — With Dmitry Shevelenko
WithDmitry Shevelenko
Questions people ask
- Why is inference becoming the strategic 'last market'?
- Companies with unique user signals can encode value into workflows and post-train specialized models, making inference a critical differentiator.
- What is Baseten's approach to GPU capacity?
- Baseten operates a multi-cloud fabric across 18 clouds and 90 clusters to address GPU constraints and serve diverse customer needs.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Conviction.