“I Was Wrong About Coding” — Jay Jung, Lead of OpenAI GTM Innovation Lab, on Building OpenAI With OpenAI
How the engineer building AI at OpenAI’s Innovation Lab rewired his thinking on coding, curiosity, and what it actually takes to stay relevant in the age of AI
Upcoming Event:
Scaling Apps to Millions: The Growth Playbook You Need | Time: Thursday, March 26, 2026 6pm-9pm | Location: Stanford | Register: https://luma.com/poy80ci3
Jay Jung is the Lead of GTM Innovation Labs at OpenAI, where his team builds zero-to-one products battle-tested internally by hundreds of OpenAI employees. One of those tools, DocGPT, sent DocuSign’s stock down 12% the day it was announced, climbing to a 17–18% drop by week’s end. (Jay’s own words: “Maybe there isn’t a direct correlation, maybe there is. I can’t really say for sure.”) His team also built the GoToMarket Assistant, aggregating knowledge across platforms like Gong, Salesforce, Ironclad, Databricks, and Tableau surfaced through ChatGPT and Slack via MCPs.
Born in Korea and raised in the US, Jay is a first-generation tech engineer who started coding in his early 20s with no family background in tech — going from 3am research sessions on Reddit and Blind to leading innovation at OpenAI in under a decade.
What We Discussed
What we discussed:
What it actually feels like to be inside OpenAI frontier AI lab, and what outsiders are missing
Why Jay hasn’t written a single line of code in over a year and what he does instead
What skills actually separate great AI builders from average ones, and why it’s not just about prompting
Whether 80% of apps are already obsolete, and what individual builders should do right now
How OpenAI uses OpenAI internally, and what that recursive loop reveals about where AI is heading
What the future of work looks like when work becomes optional, and how to prepare for it
Full Interview (YouTube):
Key Takeaways
Building OpenAI with OpenAI Is a Recursive Loop. Jay’s team doesn’t just build AI tools; they use those same tools to build better tools. As Jay put it in the interview: “We build great tools, then we use those tools to make better tools.” Codex is how his team ships products, and the faster the models get Jay hints at Codex 5.3 with Spark being roughly 10x faster (”don’t quote me”), the faster ideas go from morning thought to working product. The full implications of this compounding loop are something Jay unpacks in the video.
DocGPT didn’t just demo well; it moved markets. When Jay’s team shipped DocGPT, DocuSign’s stock dropped 12% that day and up to 17–18% by week’s end. Jay is careful to say correlation isn’t causation, but the signal is clear: frontier labs building what he calls “L+1 derivative” features on top of their own models can instantly threaten companies whose entire business is that one feature. He explains exactly how his team thinks about this and what it means for every SaaS company in the video.
Build for Now, Not for the Imagined Future. Jay’s advice to builders in the age of ephemeral apps: “You have to sort of live in the now. It’s really hard to build for the future.” First-mover advantage still matters. Someone will automate it eventually, but the ones who ship first learn fastest. He shares the exact mental model he uses to decide what to build and when to walk away.
GTM Is the New Moat, and It’s Being Rebuilt from Scratch. Jay argues that go-to-market is more important than ever, but the way it’s executed is changing entirely. “The concept of go-to-market is changing, but the importance is only amplifying.” AI SDRs, AI CSMs, bottom-up growth loops his team is building the infrastructure for all of it internally at OpenAI. He shares what GTM orgs are getting wrong right now, and the specific playbook his team is running.
References
Jay Jung(Guest): https://www.linkedin.com/in/jay1/
OpenAI Innovation LabBlogs: https://openai.com/index/building-openai-with-openai/
Jing Conan Wang(Host): https://www.linkedin.com/in/jingconan/
FounderCoHo: https://www.linkedin.com/company/foundercoho/
Timestamps
00:00 - Highlights & Intro
02:58 - Building with OpenAI: Insights and Innovations
07:03 - Inside the Tools Powering the Frontier
11:59 - The Future of Work: Embracing Flexibility
18:31 - The Art of Prompt Engineering: Skills for the Future
24:08 - Building for Now: The Transient Nature of Apps
30:30 - Go-To-Market Strategies: The Importance of Marketing
35:41 - Learning to Code: A Personal Journey
39:07 - Breaking into Tech: Challenges and Strategies
43:03 - Sharing Knowledge: The Impact of Social Media
49:51 - The Future of Work and Automation
58:24 - Inside the Frontier: Daily Life and Work Culture at OpenAI
01:04:50 - Curiosity as a Driving Force: Learning and Growth
01:12:30 - Advice to My Younger Self: Embrace AI and Curiosity
01:14:51 - The Future of Education: Personalized Learning with AI
01:19:05 - The Future of Engineering: Redefining Roles and Skills