Brett Reifers

I help teams turn agentic AI from buzzword to measurable time savings. Building agents and skills at Microsoft Digital for everyone, not just developers.

[contextmaxxing @nicholasdbrady]

@film_girl and I will be live blogging all the announcements from @mustafasuleyman and the MAI team tomorrow.

Keep this 🔗 locked: https://t.co/T6xbeVFpru

[Brett Reifers @BrettReif]

@nicholasdbrady Keep those builder insights lively!

[contextmaxxing @nicholasdbrady]

Hey @AIFoundryDevs, What's New is up for #MicrosoftFoundry last month. Big model month — GPT-5.5, GPT-image-2, Claude Opus 4.7, Gemma 4, and four first-party MAI models for image gen, voice, and transcription.

[Evan Boyle @_Evan_Boyle]

If you've been wondering what I've been up to, the team and I have been cooking up something new.

A new agent-native development environment deeply integrated with the GitHub graph. Not just for writing code, but all of the meta-work as well.

[contextmaxxing @nicholasdbrady]

agents write files, execute code, and hold credentials. if two customers are calling the same agent and it's running on a shared container, that's a security problem. not a theoretical one.

Microsoft Foundry just shipped hosted agents to fix this. public preview today. 1/

[contextmaxxing @nicholasdbrady]

Hey @AIFoundryDevs,
Big updates in Microsoft Foundry | March 2026

- Foundry Agent Service now GA: private networking, Voice Live preview & more regions
- GPT-5.4 & GPT-5.4 Pro GA: production-grade reasoning + computer use
- MAI Image-2, MAI Voice-1, MAI Transcribe-1 now available in Foundry
- Priority Processing, Phi-4 Vision, SDK 2.0 GA & more

Enterprise AI just leveled up. Full details: https://t.co/Nlv6O37wl5
#MicrosoftFoundry #AzureOpenAI

[Burke Holland @burkeholland]

When I was your age, we didn't generate slop with AI.

We googled it, found it, and pasted it in ourselves.

[VS Code Changelog @VSCodeChangelog]

It's never been easier to install plugins for skills, agents, and more. Directly from a GitHub repo or git URL.

[Brett Reifers @BrettReif]

/ftw

[Boris Cherny @bcherny]

btw

[Pierce Boggan @pierceboggan]

Folks often ask for a list of best practices partnering with agents in @code, and we've compiled a list here!

[Visual Studio Code @code]

Agents, for real work.

The latest @code release gives you better agent orchestration, extensibility, and continuity.

Here's what's new:

🪝 Hooks support
🎯 Message steering and queueing
🌐 Agentic integrated browser
🧠 Shared memory
And more...

[Pierce Boggan @pierceboggan]

"Summarizing conversation history" hates to see this setting enabled...

[SemiAnalysis @SemiAnalysis_]

4% of GitHub public commits are being authored by Claude Code right now. At the current trajectory, we believe that Claude Code will be 20%+ of all daily commits by the end of 2026. While you blinked, AI consumed all of software development.

[Tim @timheuer]

If you are using GitHub Copilot, folks you need to be watching:
@pierceboggan
@burkeholland
@_Evan_Boyle
@OrenMe
@digitarald

[Andrej Karpathy @karpathy]

It is hard to communicate how much programming has changed due to AI in the last 2 months: not gradually and over time in the "progress as usual" way, but specifically this last December. There are a number of asterisks but imo coding agents basically didn’t work before December and basically work since - the models have significantly higher quality, long-term coherence and tenacity and they can power through large and long tasks, well past enough that it is extremely disruptive to the default programming workflow.

Just to give an example, over the weekend I was building a local video analysis dashboard for the cameras of my home so I wrote: “Here is the local IP and username/password of my DGX Spark. Log in, set up ssh keys, set up vLLM, download and bench Qwen3-VL, set up a server endpoint to inference videos, a basic web ui dashboard, test everything, set it up with systemd, record memory notes for yourself and write up a markdown report for me”. The agent went off for ~30 minutes, ran into multiple issues, researched solutions online, resolved them one by one, wrote the code, tested it, debugged it, set up the services, and came back with the report and it was just done. I didn’t touch anything. All of this could easily have been a weekend project just 3 months ago but today it’s something you kick off and forget about for 30 minutes.

As a result, programming is becoming unrecognizable. You’re not typing computer code into an editor like the way things were since computers were invented, that era is over. You're spinning up AI agents, giving them tasks *in English* and managing and reviewing their work in parallel. The biggest prize is in figuring out how you can keep ascending the layers of abstraction to set up long-running orchestrator Claws with all of the right tools, memory and instructions that productively manage multiple parallel Code instances for you. The leverage achievable via top tier "agentic engineering" feels very high right now.

It’s not perfect, it needs high-level direction, judgement, taste, oversight, iteration and hints and ideas. It works a lot better in some scenarios than others (e.g. especially for tasks that are well-specified and where you can verify/test functionality). The key is to build intuition to decompose the task just right to hand off the parts that work and help out around the edges. But imo, this is nowhere near "business as usual" time in software.

[Martin Woodward @martinwoodward]

If you were not allowed to use the Copilot CLI at work because it wasn’t GA, now you can!

From Windows, WSL, Linux and Mac.

Choose your model and choose how you want to work with it. All with the same GitHub Copilot subscription.

[Brett Reifers @BrettReif]

@_Evan_Boyle Good

[Evan Boyle @_Evan_Boyle]

@eliasbchlny Autopilot and fleet are no longer experimental, in mainline!

[Evan Boyle @_Evan_Boyle]

The Copilot CLI is now GA!

[Brett Reifers @BrettReif]

@pierceboggan And attention