“Engineers are becoming sorcerers” | The future of software development with OpenAI’s Sherwin Wu
- Source
- lennysnewsletter.com
- Author
- Lenny Rachitsky
- Date

Context
Some engineering around an AI exists to compensate for weaknesses that may change with the model. Sherwin Wu, engineering lead for OpenAI's API and developer platform, describes this as a moving product-design problem. Wu's example is retrieving information: a carefully orchestrated search pipeline may sometimes give way to letting a capable model use search tools itself. Wu explicitly acknowledges that vector stores still have uses. The argument is that teams should reconsider the machinery around the model as its capabilities change, because preserving every workaround can preserve yesterday's constraints.
“95% of engineers use codecs. 100% of our PRs are reviewed by codecs.”
Sherwin Wu
“It literally feels like we're wizards casting all these spells.”
Sherwin Wu
“So I think we might actually enter into a golden age of B2B sass.”
Sherwin Wu
“The models will eat your scaffolding for breakfast.”
Sherwin Wu
“Make sure you're building for where the models are going and not where they are today.”
Sherwin Wu
- 00:00Introduction to Sherwin Wu
- 03:10AI's role in coding at OpenAI
- 06:53The future of software engineering with AI
- 12:26The stress of managing agents
- 15:07Codex and code review automation
- 19:29The changing role of engineering managers
- 24:14The one-person billion-dollar startup
- 31:40Management lessons
- 37:28Challenges and best practices in AI deployment
- 43:56Hot takes on AI and customer feedback
- 48:57Building for future AI capabilities
- 50:16Where models are headed in the next 18 months
- 53:35Business process automation
- 57:22OpenAI's ecosystem and platform strategy
- 01:00:50OpenAI's mission and global impact
- 01:05:21Building on OpenAI's API and tools
- 01:08:16Lightning round and final thoughts
podcastThe 100-person AI lab that became Anthropic and Google's secret weapon | Edwin Chen (Surge AI)
podcastWhy humans are AI’s biggest bottleneck (and what’s coming in 2026) | Alexander Embiricos (OpenAI Codex Product Lead)
podcastThe coming AI security crisis (and what to do about it) | Sander Schulhoff
podcastWhy most AI products fail: Lessons from 50+ AI deployments at OpenAI, Google, and Amazon
videoOpenAI’s head of platform engineering on the next 12-24 months of AI | Sherwin WuLenny's Podcast
podcastWhy humans are AI’s biggest bottleneck (and what’s coming in 2026) | Alexander Embiricos (OpenAI Codex Product Lead)Lenny Rachitsky
podcastAI’s third era: the rise of persistent AI coworkers | Tara Seshan (OpenAI’s product lead)Lenny Rachitsky
Checking sign-in…
Loading comments…