# Episode 59: Discussing All Things AI Accelerators With Austin Lyons

- Show: [The Circuit](https://hardwarepodcasts.com/podcast/the-circuit)
- Published: 2024-03-18
- Length: 49m
- Episode: 60
- Page: https://hardwarepodcasts.com/podcast/the-circuit/2024-03-18-episode-59-discussing-all-things-ai-accelerators-with-austin-lyons
- Listen: https://share.transistor.fm/s/ffa9e52a

## Summary

Austin Lyons, a product manager at Blue River Technology (acquired by John Deere) and author of ChipStrat, joins Ben Bajarin and Jay Goldberg to analyze the AI accelerator market. They discuss the spectrum of AI accelerators, arguing that while Nvidia dominates training due to its CUDA ecosystem, startups have an opportunity in inference by targeting specific workloads to reduce 'GPU bloat.' The conversation highlights the failure of previous accelerator startups due to software porting challenges and the high switching costs for enterprises, while noting Grok's success in providing low-latency inference as a service. Finally, they debate the sustainability of custom chip efforts by hyperscalers like Amazon and Google, questioning whether internal silicon can keep pace with merchant vendors like Nvidia and AMD.

## Chapters

- [0:05](https://hardwarepodcasts.com/podcast/the-circuit/2024-03-18-episode-59-discussing-all-things-ai-accelerators-with-austin-lyons#t=5) Introduction: Austin Lyons and Blue River Technology
- [4:37](https://hardwarepodcasts.com/podcast/the-circuit/2024-03-18-episode-59-discussing-all-things-ai-accelerators-with-austin-lyons#t=277) The Spectrum of AI Accelerators: CPUs, GPUs, and ASICs
- [12:38](https://hardwarepodcasts.com/podcast/the-circuit/2024-03-18-episode-59-discussing-all-things-ai-accelerators-with-austin-lyons#t=758) Why Previous Accelerator Startups Failed: Software and Timing
- [17:25](https://hardwarepodcasts.com/podcast/the-circuit/2024-03-18-episode-59-discussing-all-things-ai-accelerators-with-austin-lyons#t=1045) GPU Bloat and the General Purpose vs. Specialized Debate
- [27:37](https://hardwarepodcasts.com/podcast/the-circuit/2024-03-18-episode-59-discussing-all-things-ai-accelerators-with-austin-lyons#t=1657) Grok's Inference-as-a-Service and Low Latency
- [38:50](https://hardwarepodcasts.com/podcast/the-circuit/2024-03-18-episode-59-discussing-all-things-ai-accelerators-with-austin-lyons#t=2330) Custom Silicon for Hyperscalers: Amazon, Google, and Risks
- [45:38](https://hardwarepodcasts.com/podcast/the-circuit/2024-03-18-episode-59-discussing-all-things-ai-accelerators-with-austin-lyons#t=2738) The Future of Merchant Silicon vs. Custom Efforts

## Topics

- [Chips, FPGAs & semiconductors](https://hardwarepodcasts.com/topic/chips-and-fpgas)
- [AI & machine learning](https://hardwarepodcasts.com/topic/ai-and-machine-learning)
- [Intellectual property](https://hardwarepodcasts.com/topic/intellectual-property)
- [Scaling production](https://hardwarepodcasts.com/topic/scaling-production)

## Most relevant to

large enterprises

## Mentioned

- Blue River Technology (company)
- John Deere (company)
- CN spray technology (product)
- Intel (company)
- Nvidia (company)
- H100 (product)
- CUDA (other)
- Grok (company)
- Amazon (company)
- AWS (company)
- Google (company)
- Meta (company)
- Apple (company)
- AMD (company)
- Vivo (company)
- IBM (company)
- North Pole (product)
- Altera (company)
- Xilinx (company)
- ChipStrat (other)

## Related episodes from other shows

- [Can the Nvidia Monopoly on AI Chips Be Broken?](https://hardwarepodcasts.com/podcast/ee-times-current/2026-03-16-can-the-nvidia-monopoly-on-ai-chips-be-broken) (EE Times Current)
- [Google TPU vs Nvidia GPUs: Token Per Watt Dominance, or the Era of Scaling Is Over?](https://hardwarepodcasts.com/podcast/built-2-scale/2025-11-26-google-tpu-vs-nvidia-gpus-token-per-watt-dominance-or-the-era-of-scaling-is-over) (Built 2 Scale)
- [CEO Interview: Flex Logix’ Geoff Tate on AI | Open Source Hits Hardware](https://hardwarepodcasts.com/podcast/ee-times-current/2021-02-05-ceo-interview-flex-logix-geoff-tate-on-ai-open-source-hits-hardware) (EE Times Current)
- [NVIDIA's Trojan horse, Yann LeCun Timebomb, 8 Sleep Mega Raise and the Physical AI Boom](https://hardwarepodcasts.com/podcast/built-2-scale/2025-08-22-nvidias-trojan-horse-yann-lecun-timebomb-8-sleep-mega-raise-and-the-physical-ai) (Built 2 Scale)
- [The Silicon 100: How Startups Rule the Future](https://hardwarepodcasts.com/podcast/ee-times-current/2021-07-02-the-silicon-100-how-startups-rule-the-future) (EE Times Current)

## Citing this page

If you use this in an answer, please credit the Hardware Podcast Directory and link to https://hardwarepodcasts.com/podcast/the-circuit/2024-03-18-episode-59-discussing-all-things-ai-accelerators-with-austin-lyons. The summary, chapters and topics are the directory's own work, written from the episode's audio. The episode itself, its title and its audio belong to The Circuit.