A Product Market Fit Show | Startup Podcast for Founders · Mistral.vc

His startup powers OpenAI's Voice Mode. Last month, they became a unicorn. | Russ d’Sa, Co-Founder of LiveKit

·55 min·3 clips

Russ ties LiveKit’s customer pull to the “product market fit question” and jokes that the show itself should prove it.

1. A Product Market Fit Show episode centers on Russ d’Sa and LiveKit, the startup that powers OpenAI’s Voice Mode and grew from open source into infrastructure. 2. Russ d’Sa is LiveKit’s co-founder, former Twitter employee, and founder of Evie; Pablo Srivastava hosts and keeps the conversation on startup sales, acquisitions, and product-market fit. 3. The episode asks how LiveKit turned a side project into a company and how that path connects to the “product market fit question.” 4. Russ says his father worked on tech startups in the 1980s and 1990s across connectors, GPUs, and DSL, which shaped his desire to found companies. 5. He joined Y Combinator in the 2007 class, then later worked at 23andMe and Twitter before starting and selling Evie. 6. Evie reached about 14 people, raised roughly $13 million, and had about 2.5 million DAUs across its app and a Verizon distribution channel. 7. Russ says Evie could not extract enough revenue from ad networks because Verizon’s targeting was weaker than Facebook’s and the product lacked enough contextual data. 8. He describes running acquisition talks through a banker, with a plan to use strategic partnerships first and acqui-hire conversations as a hedge. 9. Coinbase, Uber, Robinhood, and Pocket/Mozilla all entered the process, and Coinbase was interested enough to interview the whole team. 10. Russ says Coinbase wanted only five people, which led him and DZ to reject that path and instead prioritize the full team. 11. Russ says the team chose a broader acquisition because they had worked together for years and did not want to optimize for an outcome that left people behind. 12. He says investors first treated the company as already zero, then became much more active once a real acquisition could return money. 13. Russ and DZ used a good-cop, bad-cop negotiation pattern so DZ could push on investors while Russ held the line on terms. 14. The Medium acquisition emerged through Josh Allman, who knew Russ from Twitter and was also on Medium’s board. 15. Russ says Ev Williams saw value in the team, the recommendation technology, and the distribution angle because Medium wanted more AI-driven content discovery. 16. LiveKit began when Russ was building a “Clubhouse for companies” and needed WebRTC infrastructure that did not exist at the API layer. 17. Russ and DZ opened a GitHub repo and Slack channel, then watched the project snowball as unknown developers asked for a React Native SDK and plugin fixes. 18. By the first public open-source release in summer 2021, LiveKit was already drawing interest from large companies, and Russ says Spotify, Oracle, X, and Reddit were all deploying it. 19. The conversation is interview-based, detail-heavy, and built around specific numbers, relationships, and negotiation stories rather than broad theory. 20. Founders selling a company, building infrastructure, or working on audio and video APIs will get the most value here; casual listeners without startup context may skip it.

As heard by us

A founder story about timing, infrastructure, and why market choice beats product polish.

It works best in the stretch that follows Russ d'Sa from a demo and a tweet that barely moved, through a five-hour lunch, a secret live-cloud test, and the later voice-mode call with OpenAI.

Read the full review in PlayNext →

Why you'd press play

Press play if you want the story behind OpenAI building voice mode on top of LiveKit.

Read the full recommendation in PlayNext →

Listen to the show on

Apple Podcasts Spotify

Play clips from this episode More from A Product Market Fit Show | Startup Podcast for Founders