Codex: The Autonomous SWE

​ChatGPT Codex just landed— a cloud-hosted Autonomous Software Engineer (A-SWE) from OpenAI.

​In this session, we’ll discover Codex, powered by codex-1

​That’s right, Codex.

​But this one is different than the Codex CLI Coding Agent OpenAI recently released and we covered in April.

​In fact, during this most recent Codex release they also announced additional Codex CLI support.

​Last month, we launched Codex CLI, a lightweight open-source coding agent that runs in your terminal. It brings the power of models like o3 and o4-mini into your local workflow, making it easy to pair with them to complete tasks faster.

​Today, we’re also releasing a smaller version of codex-1, a version of o4-mini designed specifically for use in Codex CLI. This new model supports faster workflows in the CLI and is optimized for low-latency code Q&A and editing, while retaining the same strengths in instruction following and style. It’s available now as the default model in Codex CLI and in the API as codex-mini-latest.

​~ Introducing Codex, May 16, 2025

​So the evolution of the name “Codex” continues at OpenAI, where we’ve gone from CLI command line agent to a fully autonomous software engineer.

​Here’s what we know:

• ​“Technical teams at OpenAI have started using Codex as part of their daily toolkit.”

• ​“Codex can read and edit files, as well as run commands including test harnesses, linters, and type checkers.”

• ​“Once Codex completes a task, it commits its changes in its environment.”

​In this hands-on session, we discuss exactly how powerful this tool really is, and how we as aspiring AI Engineers should think about leveraging tools like this in our workflows.

​Additionally, we’ll discuss use cases for A-SWEs like Codex, and gain some insight into why people are saying

​“Codex isn’t just a better code model. It’s an agent with its own computer, a sandboxed terminal, and a playbook for writing, testing, and landing PR-ready code—all while you sleep.” — ChatGPT Codex: The Missing Manual

​Traditional AI coding tools autocomplete your thoughts; Codex executes them. By giving the model a dedicated cloud environment, persistent filesystem, and sandboxed shell, OpenAI unlocks hours-long reasoning loops, automated test runs, and self-generated pull-requests that arrive ready to merge.

​Consider an example. Your next sprint could look like this:

1. ​Kick off up to 60 parallel Codex instances per hour to explore ideas or tackle flaky tests.

2. ​Let Codex draft and refine an Agents.md hierarchy that captures architecture guidelines, code style, and repo-specific conventions.

3. ​Watch it author concise PR descriptions, link to deterministic test logs, and cite every file it touched—no bespoke prompt engineering required.

4. ​Approve or tweak; Codex rebases, retests, and ships.

​Join us in live to learn how to set up workflows like this one.

​🧭 What you’ll learn

• ​The architecture behind Codex’s cloud “personal computer”

• ​Best-practice playbook: linters, commit hooks, modular design & abundance mindset prompting

• ​How Agents.md and smart repo naming super-charge agent discoverability

• ​Scaling tips for concurrent agents and long-running jobs

• ​Where Codex sits in the broader agentic tooling landscape (Devin, Claude Code, Cursor, etc.)

​🤖 Who should attend

• ​AI/ML engineers building agent-oriented dev workflows

• ​Platform teams evaluating automated PR pipelines

• ​Tech leads hunting for 10× engineer leverage without 10× headcount

• ​Anyone curious about the future of autonomous software engineering

​Speakers:

• ​Dr. Greg” Loughnane is the Co-Founder & CEO of AI Makerspace, where he is an instructor for their AI Engineering Bootcamp. Since 2021, he has built and led industry-leading Machine Learning education programs.  Previously, he worked as an AI product manager, a university professor teaching AI, an AI consultant and startup advisor, and an ML researcher.  He loves trail running and is based in Dayton, Ohio.

• ​Chris “The Wiz” Alexiuk is the Co-Founder & CTO at AI Makerspace, where he is an instructor for their AI Engineering Bootcamp. During the day, he is also a Developer Advocate at NVIDIA. Previously, he was a Founding Machine Learning Engineer, Data Scientist, and ML curriculum developer and instructor. He’s a YouTube content creator YouTube who’s motto is “Build, build, build!” He loves Dungeons & Dragons and is based in Toronto, Canada.

​Follow AI Makerspace on LinkedIn and YouTube to stay updated about workshops, new courses, and corporate training opportunities.