Discover how Codex unifies IDE, CLI, GitHub, mobile and cloud into a single developer agent. Practical workflows, demos, and repo patterns to make Codex a productive teammate.

Codex Everywhere: One Unified Agent for IDE, CLI, Cloud & GitHub

Build Hour: Codex — One Agent for Everywhere You Code

Codex is now one agent for everywhere you code — connected by your ChatGPT account. This Build Hour is a hands-on walkthrough of how to use all its features, including the new IDE extension and code review.

Dominik Kundel (Developer Experience) and Pranav Deshpande (Product Marketing) cover:
- What’s new with Codex? IDE extension, revamped Codex CLI, code review, and local to cloud handoffs
- How Codex works: where you can use it, and where it runs
- Live demos for pair programming with Codex CLI and IDE extension
- Best practices for structuring your codebase and delegating tasks to the - Codex cloud agent
- Live Q&A

👉 Follow along with the code repo: https://github.com/openai/build-hours
👉 Codex docs: https://developers.openai.com/codex
👉 Agents.md: https://agents.md/
👉 Codex CLI repo: https://github.com/openai/codex
👉 Sign up for upcoming live Build Hours: https://webinar.openai.com/buildhours/

Build Hour: Codex

An HR pro reveals how Canva, imaginative design, and soft-skill exercises transformed dull slide decks into memorable, actionable presentations.

From Snooze to Spellbound: How Canva Made My HR Presentations Magical

Canva | Spellbound: How I Turned My HR Presentations Into Real Magic

Meet Wendy, an HR professional whose presentations could use a bit of charm. Watch her put imagination to work and bewitch the crowd with Canva.

Wendy’s HR presentations were more snooze than spellbinding until imagination cast its charm.

Canva | Spellbound

See how a thriller-inspired rebrand made with Canva turned Bob’s quiet ice cream truck into a local sensation—behind-the-scenes, step-by-step workflow, and results for small businesses.

Vamped Up: A Thriller Rebrand for Bob's Ice Cream Truck

Canva | Vamped Up: How I Helped Bob Turn Plain Vanilla Into Pure Thriller

Meet Bob, an ice cream man whose social posts are as plain as vanilla. Watch him put imagination to work and vamp up his content with Canva.

Zero customers. One ice cream truck. And an imaginative rebrand you won’t see coming. Bob’s about to turn plain vanilla into pure thriller.

Canva | Vamped Up

See how Barb's playful redesign of a flat realtor sign turned heads, sparked conversation, and proves small design choices can boost local marketing.

DIY 'Slayyy' Sign: How One Makeover Sparked Realtor Buzz

Canva | Slayyy — How One Flat Sign and Barb's Imagination Transformed a Realtor's Message

One realtor. Flat signs. Zero vibe. Then Barb’s imagination stepped in… and everything changed.

Canva | Slayyy

Respectful, detailed reporting on the Sonderkommando at Auschwitz-Birkenau. Survivor testimonies expose the brutal mechanics of the crematoria and the moral complexity of forced survival.

Inside the Sonderkommando: Testimonies from Auschwitz

Sonderkommanda of Auschwitz: The Most Terrifying Work in History

V samotném srdci Osvětimi existovala skupina vězňů donucených vykonávat ten nejstrašnější úkol v celém systému vyhlazování: Sonderkommanda. Byli to muži nucení pracovat v krematoriích a spalovacích jámách – tahali čerstvě zavražděná těla z plynových komor, ve dne v noci je vhazovali do pecí a drtili kosti, aby zahladili jakékoli stopy po obětech. Pod neustálým dohledem SS neměli na výběr: přežít znamenalo plnit ten nejnehumánnější úkol, jaký nacistický režim vymyslel.

Svědectví přeživších, jako Filip Müller, Henryk Tauber nebo Szlomo Dragon, popisují nesnesitelné scény: dusivé horko z pecí, zápach spáleného masa pronikající vzduchem, přímý kontakt s ještě teplými těly a prach z kostí pokrývající kůži i plíce. Nebyly žádné přestávky ani oddech. Práce byla nepřetržitá, mechanická a vykonávaná pod hrozbami a bitím. Sonderkommanda navíc věděla, že jejich vlastní osud bude nakonec stejný jako osud obětí: po několika týdnech či měsících byli nahrazeni a popraveni, aby nezůstal žádný svědek.

Navzdory útlaku někteří dokázali zanechat stopy svých zážitků. Rukopisy zakopané poblíž krematorií, pozdější výpovědi a povstání z října 1944 ukazují, jak tito muži, uprostřed hrůzy, usilovali o to, aby zůstalo svědectví o tom, co viděli. Jejich příběh je tím nejtemnějším z Holokaustu: lidské bytosti redukované na součástky vyhlazovací mašinérie, odsouzené nést v paměti tu nejstrašnější zločinnost v dějinách.

Sonderkommanda z Osvětimi: Nejděsivější práce v dějinách

Learn how a 'ComfyUI, AI, Educational' approach and GPT-5-powered, domain-specific assistants (nearly 25,000 custom GPTs) help Bain accelerate client value, govern AI, and measure ROI.

Scaling Consulting Impact with GPT-5: Bain’s ComfyUI Playbook

How Bain Accelerates Client Value with GPT-5 — A Deep Dive into ComfyUI, AI, Educational, Approaches for Moving from Plan to Action

Bain & Company has embedded GPTs across its business, with dozens of proprietary applications and nearly 25,000 custom versions in use. Gene Rapoport, Partner & AI Leader in the Private Equity practice, shares how GPT-5 drives better outputs — and improved client results — by powering everything through one consistent, flexible model.

Bain accelerates client value with GPT-5

Explore agentic coding with Claude Code: practical scaffolding, ComfyUI-style workflows, evaluation methods, and hands-on tips to safely automate repo changes.

Agentic Coding Explained: Practical Claude Code Workflows

The Future of Agentic Coding: Claude Code, Practical Advice, and a ComfyUI, AI, Educational, Approach

Anthropic's Boris Cherny (Claude Code) and Alex Albert (Claude Relations) discuss the current and future state of agentic coding, the evolution of coding models, and designing Claude Code's "hackability." Boris also shares some of his favorite tips for using Claude Code.

0:00 - Introductions
0:39 - The current state of agentic coding
5:20 - The evolution of coding models
7:39 - Coding model evaluation 
8:56 - Claude Code user feedback loops
10:34 - The “hackability” of Claude Code (CLAUDE.md, MCP, slash commands)
13:11 - The future of agentic coding
14:49 - How to upskill for agentic coding
17:49 - Claude Code tips and tricks

Learn more about Claude Code: http://clau.de/future-of-agentic-coding
Check out the Claude Code docs: https://clau.de/claude-code-docs

The future of agentic coding with Claude Code

Recap of the ImageGen Build Hour: GPT-native image generation, Responses API features, streaming previews, multi-turn edits, demos, customer spotlight and best practices.

ImageGen Deep Dive — GPT‑Native Image API, Streaming & Multi‑Turn Editing

Build Hour: Image Gen

Over 130M people created images with Image Gen in its first week inside ChatGPT. Now, with Image Gen in the API, developers can build the same capabilities directly into their own apps and platforms.This Build Hour walks through gpt-image-1 in the API, with demos on streaming, editing, and masking for real-world apps.

Bill Chen (Applied AI) covers:
- What’s new: text rendering, world knowledge, image inputs
- New capabilities: streaming, multi-turn editing, masking
- Best practices: picking the right API (image vs responses), customizing outputs, handling latency & UX tradeoffs 
- Live demo: building an AI-powered photobooth from scratch
- Customer spotlight: create AI presentations using Gamma (https://gamma.app/)
- Live Q&A

👉 Follow along with the code repo: https://github.com/openai/build-hours
👉 Check out the Image Gen Guide: https://platform.openai.com/docs/guides/image-generation
👉 Sign up for upcoming live Build Hours: https://webinar.openai.com/buildhours

Build agentic systems with OpenAI's 2025 toolset—Responses API, Agents SDK, hosted tools, and Codex. A hands-on guide with demo architecture, evaluation tips, and product UX advice.

Agentic Tool Calling: Build Goal-Directed Agents with the Responses API

Build Hour: Agentic Tool Calling

In 2025, agents don’t just think — they run code, call tools, and complete tasks.

This Build Hour is a hands-on walkthrough of how to design agentic systems that reason and act using OpenAI’s latest APIs and SDKs. 

Ilan Bigio (Developer Experience) covers:
- What’s new in 2025: Responses API, Agents SDK, Hosted Tools, Codex, and more
- Chain of thought concepts: reasoning, tool calling, and long-horizon tasks
- Live demo: building an agentic task system to process a backlog of tickets
- Delegation: directional guidance for evals
- Live Q&A

👉 Follow along with the code repo: https://github.com/openai/build-hours
👉 Check out additional resources: https://developers.openai.com/
👉 Sign up for upcoming live Build Hours: https://webinar.openai.com/buildhours

A practical guide to reinforcement fine-tuning (RFT): why it beats traditional fine-tuning for reasoning tasks, sample-efficient workflows, demo insights, and a hands-on checklist.

Reinforcement Fine-Tuning for LLMs — Boost Reasoning & Domain Accuracy

Build Hour: Reinforcement Fine-Tuning

Reinforcement fine-tuning (RFT) lets you improve how models reason by training with graders instead of large labeled datasets. This Build Hour shows you how to set up tasks, design grading functions, and run efficient training loops with just a few hundred examples.

Prashant Mital and Theophile Sautory (Applied AI) cover:
- Intro to RFT: optimization, fine-tuning options, RFT benefits
- Task setup: prompts, graders, and training and validation data
- Live demo: building and running RFT for a classification task
- RFT workflow: from dataset selection to evaluating and iterating
- Customer spotlight: Accordance uses RFT models for tax and accounting workflows (https://accordance.com/)
- Live Q&A

👉 Follow along with the code repo: https://github.com/openai/build-hours
👉 RFT Cookbook: https://cookbook.openai.com/examples/reinforcement_fine_tuning
👉 RFT Use Case Guide: https://platform.openai.com/docs/guides/rft-use-cases
👉 Sign up for upcoming live Build Hours: https://webinar.openai.com/buildhours

Recap of OpenAI Build Hour: how real-time speech-to-speech models and new SDKs are reshaping voice agents — architectures, demos, and production best practices for deployers.

Real-Time Speech-to-Speech: The Next Wave in Voice Agents

Build Hour: Voice Agents — How Real-Time Speech-to-Speech Models Are Changing Conversational AI

Voice agents don’t just transcribe anymore — they think, talk, and call tools in real time.

This Build Hour demos speech-to-speech agents built with the Realtime API and Agents SDK that can handle conversations natively in audio, reason about context, and call tools while streaming speech back to the user.

Brian Fioca and Prashant Mital (Applied AI) cover:
- Why voice agents now: APIs to the real world, expressive + accessible interactions
- Architectures: chained speech-to-text vs. end-to-end speech-to-speech models
- Live demo: building a voice-powered workspace manager + designer agent with handoffs
- Best practices: evals, guardrails, and delegation
- Live Q&A

👉 Follow along with the code repo: https://github.com/openai/build-hours
👉 Check out the voice agents guide: https://platform.openai.com/docs/guides/voice-agents
👉 Sign up for upcoming live Build Hours: https://webinar.openai.com/buildhours

Build Hour: Voice Agents

Step-by-step developer guide to integrating Gemini Nano‑Banana (Gemini 2.5 Flash Image). Learn AI Studio prototyping, API key setup, SDKs, photorealistic image generation & editing, restoration, and conversational workflows.

Gemini Nano‑Banana Image API — Practical Developer Guide

Learn to Build with Gemini Nano-Banana (Gemini 2.5 Flash Image)

Learn how to build with Nano Banana in this complete developer tutorial. This video provides a comprehensive walkthrough for developers looking to integrate Nano Banana aka Gemini 2.5 Flash Image into their applications.

Resources:
Blog post →https://goo.gle/4pozDUu 
Use Nano Banana in AI Studio → https://goo.gle/4gdLiRK 
JavaScript examples →https://goo.gle/4p5yhhj 
Nano Banana Hackathon → https://goo.gle/45XWbCk 

Timeline:
00:00 - Introduction
00:32 - AI Studio
01:25 - Project Setup
03:16 - Image creation
05:47 - Image editing
06:58 - Multiple input images
07:35 - Photo restoration
08:12 - Conversational image editing
09:32 - Best practices and effective prompting

Subscribe to Google for Developers → https://goo.gle/developers 

Speaker: Patrick Lober
Products Mentioned: Google AI, Gemma, Gemini

AI and Technology News

Latest