🤖 AI Summary
Overview
This episode dives into OpenAI's Codex, exploring its groundbreaking speed, the strategic shift towards a broader audience, and the development of its GUI-based app. Thibault Sottiaux and Andrew Ambrosino discuss the evolution of Codex, its applications, and how it’s reshaping coding workflows. They also share insights into the challenges of building and optimizing Codex, as well as its positioning against competitors.
Notable Quotes
- The first time I showed it to someone, they were like, 'No way. This is a fake demo. This cannot be this fast.'
– Thibault Sottiaux, on the revolutionary speed of Codex.
- GUIs are great, IDEs are just the problem. There's something that's a GUI for programming that's not an IDE, and it seems like you're figuring that out.
– Dan Shipper, reflecting on the Codex app's unique approach.
- It’s like a command center for your agent, optimized around the idea that you have a very capable, intelligent entity that you’re controlling, steering, and supervising.
– Thibault Sottiaux, on the vision for Codex.
🚀 Strategic Shift in Codex's Audience
- OpenAI's decision to run a Super Bowl ad for Codex marked a shift from targeting professional engineers to appealing to a broader audience.
- The Codex app, with its user-friendly interface and multitasking capabilities, aims to inspire non-technical users to explore coding and agent-based workflows.
- Codex 5.3, a highly capable and fast model, reinforces OpenAI's commitment to making coding accessible and productive for diverse users.
🖥️ GUI vs. Terminal: A Bold Design Choice
- The Codex team prioritized a GUI over a terminal-based interface (TUI), believing it offers a more intuitive and flexible experience for users.
- Features like voice prompting, mermaid diagrams, and persistent panes enhance the GUI's functionality, making it a command center
for coding and agent management.
- The app's design reflects a shift away from traditional IDEs, focusing instead on a streamlined, dynamic interface tailored to modern coding workflows.
⚡ Speed as a Game-Changer
- Codex 5.3's unprecedented speed allows users to stay in the flow, reducing the need for multitasking and enabling real-time interactions.
- The model's speed opens up new possibilities, such as mid-turn steering, where users can adjust tasks dynamically while the model is working.
- Faster models also improve end-to-end testing and integration, making workflows more efficient and seamless.
🤖 Automations and Skills in Codex
- Codex's automation capabilities enable users to schedule tasks, resolve merge conflicts, and generate reports, significantly enhancing productivity.
- Unique use cases include creating personalized books, automating marketing research, and even fixing subtle bugs in random files.
- Skills like Image Gen
and Yeet
streamline workflows, showcasing the app's versatility beyond traditional coding tasks.
🛠️ Challenges and Future Bottlenecks
- Verification of generated code remains a significant bottleneck, as humans still need to ensure correctness and consistency.
- The team is exploring ways to integrate models into code review processes, using features like annotated diffs and automated testing evidence.
- OpenAI aims to address these challenges by leveraging Codex's speed and intelligence to make verification and understanding more efficient.
AI-generated content may not be accurate or complete and should not be relied upon as a sole source of truth.
📋 Episode Description
OpenAI’s hottest app isn’t ChatGPT—it’s Codex.
In the last few weeks alone, the Codex team shipped a desktop app, GPT-5.3 Codex (a new flagship model), and Spark, the fastest coding model I’ve ever used. Usage has grown fivefold since January, and over a million people now use Codex weekly. Codex was also the app that OpenAI chose to run an ad for in the Super Bowl.
Dan Shipper talked to Thibault Sottiaux, head of Codex, and Andrew Ambrosino, a member of technical staff who built the Codex app, for Every’s AI & I about what OpenAI is building and how they’re using it internally.
If you found this episode interesting, please like, subscribe, comment, and share!
Want even more?
Sign up for Every to unlock our ultimate guide to prompting ChatGPT here: https://every.ck.page/ultimate-guide-to-prompting-chatgpt. It’s usually only for paying subscribers, but you can get it here for free.
To hear more from Dan Shipper:
Subscribe to Every: https://every.to/subscribe
Follow him on X: https://twitter.com/danshipper
Head to granola.ai/every and get 3 months free with the code EVERY.
Timestamps:
00:00:00 - Start
00:01:27 - Introduction
00:05:27 - OpenAI's evolving bet on its coding agent
00:09:42 - The choice to invest in a GUI (over a terminal)
00:20:38 - The AI workflows that the Codex team relies on to ship
00:26:45 - Teaching Codex how to read between the lines
00:28:45 - Building affordances for a lightening fast model
00:33:15 - Why speed is a dimension of intelligence
00:36:30 - Code review is the next bottleneck for coding agents
00:41:24 - How the Codex team positions against the competition
Links to resources mentioned in the episode:
Thibault Sottiaux: Tibo (@thsottiaux)
Andrew Ambrosino: Andrew Ambrosino (@ajambrosino)
Every’s vibe check on everything the Codex team launched: OpenAI's Codex App Gains Ground on Claude Code, GPT-5.3 Codex—The 10x Engineer, Now More Fun at Parties, AI as Fast as Your Train of Thought