Title slide: One Developer, Two Dozen Agents, Zero Alignment: Why we Need Collaborative AI Engineering

The meme of two astronauts looking at earth. The one behind has a gun pointed at the other's head. First says "wait, alignment is the bottleneck?". Second says "always has been"

Putting together the final touches on slides for my talk at AI Engineer conf this Thursday @aidotengineer.bsky.social

Talking about how we need realtime, multiplayer, collaborative agentic dev tools. Not isolated, single player terminal instances.

And showing one we've built at @githubnext.com!

That's why it's a stylistic set — you can turn them on or off according to your preferences!

View Agentic Workflow configs in the Actions run summary - GitHub Changelog You can now view your Agentic Workflow markdown configs directly in the GitHub Actions run summary for any agentic workflow. This update makes it easier to review runs by: Reducing…

You can now see AW files directly in actions runner page - github.blog/changelog/20...

For the keen observers among you — the case study references Greek and Cyrillic support which is done but waiting on us to roll the 1.4 release. Coming very very soon!

Lettermatic ✕ GitHub Next: the inside story of Monaspace, a truly groundbreaking advancement for how we display code. We knew that Lettermatic would go deep, but we were still amazed by their level of knowledge and craft.

See their case study below, and get the fonts at monaspace.githubnext.com

Suddenly, the future arrived?

❤️❤️❤️ love to see it! Big update coming soon for monaspace 🔡

🌈Repo Assist: A Repository Assistant I have spent the weekend creating and using a preliminary version of what I call Repo Assist, which is an automated repository assistant for software maintainers. For some repos, some of the time. …

My latest blog post is out!

🌈 Repo Assist: A Repository Assistant

Repo Assist is an experimental automated assistant for software maintainers. For some repos, not all.

It is a GitHub Agentic Workflow and uses Copilot, Claude or Gemini in automated, proactive mode.

dsyme.net/2026/02/25/r...

Much like @steve-yegge.bsky.social's 8 levels of AI adoption, the superusers of Agentic Workflows are not just doing the same work as before, they're moving the conventions around what is automatable.

It's wild to see this firsthand inside our team. ❤️

Coda: You can get value out of this in five minutes, but getting maximum value requires a behavioral shift, and folks at Next are buzzing about this.

This tool can do many things. It's easy to use, but harder to realize/internalize that you can apply it to problems you wouldn't have before.

And we're particularly excited about using this to help open-source maintainers. When we showed this to @frenck.social, the maintainer of Home Assistant, he wrote a workflow that reads the traces for bugs and figures out if they originate in the HA codebase or not.

True noise elimination for OSS.

We're seeing teams at GitHub use it for building tooling and workflows for themselves.

We're seeing developer-adjacent personas at GitHub using agentic workflows to help them move information around.

Software is a team sport and this helps everyone on the team, not just the engineers.

You can check out a zillion examples in our docs site:

gh.io/aw-tp-bsky

It can be for easy things like "triage my issues" and it can be for hard things like "find ways to optimize my code".

If it's a task you'd express in English to an engineer, it might be a good agentic workflow!

Core to the behavior of Agentic Workflows is the notion of safe outputs. The body of a workflow is markdown, but the frontmatter is where you explicitly define what the agent can read and write, and those are enforced outside of the agentic run.

Heuristics can only take you so far in automation. There's no linter in existence that can tell you if your code and documentation have drifted out of sync.

Agentic Workflows make it stupidly simple to express a task you want to delegate, forever — and have it do that for you, forever.

Automate repository tasks with GitHub Agentic Workflows Build automations using coding agents in GitHub Actions to handle triage, documentation, code quality, and more.

We're super jazzed to be launching GitHub Agentic Workflows into technical preview today!

"Generative AI" kind of fooled us into thinking that AI is synchronous, but async AI can have so much more value for us as developers. 🧵

github.blog/ai-and-ml/au...

We've just shipped something fun - Technical Preview of GitHub Agentic Workflows by @githubnext.com

Check out blog post from @dsyme.bsky.social and Peli de Halleux

github.blog/ai-and-ml/au...

7 learnings from Anders Hejlsberg: The architect behind C# and TypeScript Anders Hejlsberg shares lessons from C# and TypeScript on fast feedback loops, scaling software, open source visibility, and building tools that last.

Why did TypeScript work when so many languages didn't?

Anders Hejlsberg, the mind behind Turbo Pascal, Delphi, C#, and TypeScript shares his thoughts and learnings.

I've been lucky enough to build with the @githubnext.com team for the last few months and I can't wait to share what we've been dreaming up at Universe! ✴︎

Also I'll be there at the Next booth so come say hi 👋

GitHub Universe 2025 session 'From promise to practice: The future of agentic collaboration'. Features GitHub Next team discussing how AI agents will transform software development, with speakers Idan Gazit, Russell Horton, Eddie Aftandilian, and Terkel Gjervig.

Going to be at Universe 25? Come see what the Next team has been cooking 🍳

The Github Next website: "Github Next investigates the future of software development"

First day of my new life at Github Next! @githubnext.com ✨

I've had my eye on this team for a long time, and it feels like the perfect moment to join them.

Join the GitHub Next Discord Server! A place to connect with the people working at GitHub Next, and other folks thinking about the future of software dev. | 12590 members

6/ 💬 Join the conversation in the Continuous AI channel on the GitHub Next Discord: gh.io/next-discord

GitHub - githubnext/agentics: Samples for GitHub Agentic Workflows Samples for GitHub Agentic Workflows. Contribute to githubnext/agentics development by creating an account on GitHub.

5/ 🔍 This is a research demonstrator, not a product or even a prototype. We’re learning in the open. Help us figure out what works, what’s safe, and what’s reusable!

Examples: github.com/githubnext/agentics

4/ ✨ Key ideas:
• GitHub-native, Actions-first
• Engine-neutral (swap LLMs without rewriting)
• Shareable workflow pieces
• Strong guardrails + auditable execution

3/❓Why?

Some tasks are perfect for natural language:
• Continuous documentation updates
• Issue triage
• Accessibility reviews
• Test improvements
• Continuous QA

All repetitive, collaborative, and needing a bit of judgment.

A GitHub workflow YAML file depicting an example agentic workflow

2/ 🤖 With Agentic Workflows, you can:
• Write repo automation in Markdown
• Compile it to Actions YAML
• Run it on GitHub using engines like Claude Code or OpenAI Codex
• Keep everything auditable, source-controlled, and team-friendly

GitHub - githubnext/gh-aw: GitHub Agentic Workflows from GitHub Next GitHub Agentic Workflows from GitHub Next. Contribute to githubnext/gh-aw development by creating an account on GitHub.

1/ 🚀 New from Next: Agentic Workflows, a research demonstrator exploring natural-language programming for GitHub Actions.

It’s one concrete manifestation of what we’re calling “Continuous AI,” focusing on repo-level automation you can describe in plain English.

🔗 github.com/githubnext/gh-aw

A “department of fool around and find out" – that's how #GitHubNext approaches innovation!

In this #InfoQ #podcast, @gazit.me & @eaftan.bsky.social reveal how their team drives the future of #AI -powered developer tools through rapid experimentation & user feedback.

🎧 Listen now: bit.ly/3UfCv7w

GitHub Next | Project Copernicus GitHub Next Project: Exploring LLM-powered navigation for your codebase

I've published a project page about small exploration I've done earlier this year around code navigation / visualization of code structure using LLMs.

Feel free to check it out:

githubnext.com/projects/cop...

For the last 10 years I’ve been wanting a chance to design the future—not tomorrow, or the next thing, but 3 things after that. The 5 year horizon.

To explore design fiction, maybe even a bit of science fiction?

Tomorrow, I get to give it a shot with my first day @githubnext.com !