🎧 How OpenAI’s Codex Team Uses Their Coding Agent

2026-02-18 · Source: AI & I - Every · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Emerging Technologies & Innovation · Depth: Advanced, extended

Summary

OpenAI recently released a new episode of its podcast "AI & I," featuring Thibault Sottiaux, head of Codex, and Andrew Ambrosino, a member of the technical staff on the Codex app. The discussion highlights the strategic shift for Codex, moving beyond professional developers to a broader, technically-adjacent audience, as evidenced by a Super Bowl ad. Since February, the Codex team has launched a desktop app, the GPT-5.3 Codex flagship model, and a research preview of a significantly faster model, GPT-5.3-Codex-Spark. This momentum has led to a fivefold increase in usage since the start of the year, with over a million weekly users. The conversation also delves into OpenAI's product strategy, the decision to prioritize a graphical user interface (GUI) over a terminal, and the team's internal workflows using automations and skills within the Codex app.

Key takeaway

For AI Engineers and technical builders evaluating coding agents, OpenAI's Codex app, with its new GPT-5.3 Codex model and the ultra-fast GPT-5.3-Codex-Spark, offers a dedicated GUI experience that prioritizes workflow integration and speed. You should explore its automation and custom skill features to streamline development tasks and consider how its rapid iteration capabilities could shift your approach to code verification and review, potentially making traditional code review less critical by directly verifying outcomes.

Key insights

OpenAI's Codex is expanding its user base and capabilities, emphasizing speed and a dedicated GUI for enhanced developer experience.

Principles

• Dedicated experiences improve developer workflow.
• Speed is a dimension of intelligence.
• AI should intuit user intent beyond literal instructions.

Method

The Codex team uses "automations" for scheduled tasks (e.g., resolving merge conflicts, daily reports) and "skills" to connect to external tools, enabling complex workflows beyond code generation, such as marketing research and image generation.

In practice

• Automate merge conflict resolution for PRs.
• Generate daily team digests of code contributions.
• Use AI for random bug hunting in codebases.

Topics

• OpenAI Codex
• Coding Agents
• AI Workflow Automation
• Model Speed Optimization
• Code Review

Best for: AI Engineer, Machine Learning Engineer, Software Engineer, AI Product Manager

Related on AIssential

• Codex + GPT-5.5 = SUPER APP! Build and Do ANYTHING! — OpenAI's GPT 5.5 and Codex offer a superior, more efficient AI coding and automation workflow.
• OpenAI Codex App: death of the VSCode fork, multitasking worktrees, Skills Automations — Agent-native coding interfaces and specialized open models are rapidly advancing, shifting developer workflows and evaluation paradigms.
• OpenAI named a Leader in enterprise coding agents by Gartner — OpenAI's Codex, powered by GPT-5.5, leads enterprise AI coding agents by enabling agentic software development with robust governance and security.
• The Codex App Has Upended My Daily Workflow — The Codex app transforms coding into managing AI agents, enabling parallel work and sustained context through a desktop interface.
• GPT-5.3 Codex Is INSANE! OpenAI’s BEST Model Might Beat Opus 4.6? (Fully Tested) — GPT-5.3 Codex offers enhanced speed, autonomy, and full software lifecycle support for coding and knowledge work.
• 5 Fun Projects Using OpenAI Codex — OpenAI Codex accelerates software development across diverse projects, acting as a versatile coding partner.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by AI & I - Every.