11 tracked signals on ai-coding-agents.
Scale AI coding agents while controlling costs with ug – the Unity Gateway CLI
Databricks · Sep 24, 2026
Databricks' Unity Gateway CLI lets enterprises govern and cost-control AI coding agents like Claude Code and Codex.
“From 0 to 60% I use Claude Code and the models are available there. 60 to 80% of the time I use Codex and a specific system model. And if I'm hitting 80-100% of my budget, I want to use a cheaper model like Kimmy.”
How SWE-Serve Exposes the Gap Between Local Tests and Live Serving
NVIDIA Developer Blog · Sep 23, 2026
NVIDIA's SWE-Serve benchmark tests whether AI coding agents' patches work in live model serving, not just local tests.
“An AI coding agent’s patch can pass tests yet fail when the server loads a real model and handles requests.”
Open Source Is Dead. Long Live Open Source. — Saoud Rizwan, Cline
AI Engineer · Aug 07, 2026
AI-generated contributions are eroding trust and community in open source projects.
“GitHub is effectively uh an archive of SLOP PRs and issues and security reports um where the sense of community before has turned into this like deep skepticism and distrust”
How Forward Deployed Engineering is done at Cognition — Jia Wu
AI Engineer · Jul 28, 2026
Cognition shipped nearly 10x more PRs internally by dogfooding Devin over six months.
“using our agent, we were able to ship almost an order of magnitude more good quality robust PRs across the organization”
Ensuring your code works when AI testing isn’t enough | BRK208
Andrej Karpathy · Microsoft Developer (Build) · Jun 04, 2026
Professional AI-assisted coding needs a new term—'agentic engineering'—distinct from casual 'vibe coding,' with verification as the core challenge.
“when you fully give into the vibes, embrace exponentials, and forget that the code even exists”
Use open weight models as your AI coding agent with Amazon Bedrock
AWS Machine Learning Blog · Sep 23, 2026
Open weight models on Amazon Bedrock let developers run private, pay-per-use AI coding agents like OpenCode.
“What if you could run an AI coding agent that keeps your data in your own AWS account, switches between frontier open weight models on demand, and charges only for what you consume?”
Writing Safe Code with AI Security and Accessibility Checks
Microsoft Developer (Build) · Jun 24, 2026
Microsoft engineers built reusable security, accessibility, and PR-review agents that validate AI-generated code before it ships.
“There may be moments where you feel like you're vibe coding it, but when you're doing it for a real product, you often have that lens of making sure that the code is secure and good performing.”
From CLI to PR: Automating the path to merged code | BRK203
Microsoft Developer (Build) · Jun 11, 2026
GitHub's CLI streamlines the AI-assisted path from issue to merged pull request.
“But now we're hitting about 275 million per week and that was just the last number I heard.”
Build fast, not fragile with Rayfin and Microsoft Fabric | OD810
Microsoft Developer (Build) · Jun 03, 2026
Microsoft introduces Rayfin, a managed backend-as-a-service on Fabric to take vibe-coded prototypes safely to production.
“multiple surveys and analyses uh show that 80 to 90% of AI generated or vibe coded prototypes never become production apps”
Running Python ASGI apps in the browser via Pyodide + a service worker
Simon Willison · May 30, 2026
Simon Willison ran Python ASGI apps in-browser via Pyodide and Service Workers, fixing prior JavaScript execution limits.
“of figuring out how to run Python ASGI apps in Pyodide using Service Workers instead, and it seems to work!”
WWDC26: Build, deliver, and automate with Xcode Cloud | Apple
Apple Developer (WWDC) · Jun 08, 2026
Apple refined Xcode Cloud's build, test, and distribution flows as AI agents in Xcode drive faster code output.
“With support for agents arriving in Xcode, a lot of developers are writing more code than ever before, iterating quickly on features and changes, and delivering more value to users”