via @dotnet : Modernize .NET Anywhere with GitHub Copilot

https://ift.tt/5x4R9PK
#dotnet #NET #GitHubCopilot #Copilot #modernize-dotnet #modernize # modernization #VSCode #VisualStudio #GitHubCLI #CLI #codingagents #customskills #developerexperience #softwaredev…

"We don't have time." "No QA engineer." "Tests were too flaky." "It's on the roadmap."

We looked at why teams skip E2E test automation. The blockers are almost never technical — they're organizational.

We wrote up the 5 most common ones and what actually helps:

https://www.kiteto.ai/en/blog/e2e-test-automation-barriers?utm_source=mastodon&utm_medium=social&utm_campaign=kw11&utm_content=e2e-blockers

#TestAutomation #E2ETesting #QA #SoftwareDev

Extend your coding agent with .NET Skills | by Tim Heuer

https://devblogs.microsoft.com/dotnet/extend-your-coding-agent-with-dotnet-skills/?hide_banner=true

#dotnet #ai #agents #skills #csharp #githubcopilot #softwaredev

Air Launches as Public Preview – A New Wave of Dev Tooling Built on 26 Years of Experience

https://blog.jetbrains.com/air/2026/03/air-launches-as-public-preview-a-new-wave-of-dev-tooling-built-on-26-years-of-experience/

#jetbrains #ai #softwaredev #agents #devtools

Why AI coding tools still need senior engineers:

→ AI writes confidently — even when wrong
→ Can't own architecture or trade-off decisions
→ Junior + AI = faster mistakes in production
→ Senior judgment = AI as leverage, not liability

Speed without correctness isn't a feature.

It's a risk.

#AIEngineering #SoftwareDev #CTOLife

Fundamental tools for software dev and systems config are quickly getting infected with AI vomit. I’m just gonna have to leave the industry, aren’t I?

Fuck all of this.

#AntiAI #SoftwareDev

Nobody gets promoted for simplicity: https://terriblesoftware.org/2026/03/03/nobody-gets-promoted-for-simplicity/

#softwaredev

Why do programmers prefer dark mode? Because light attracts bugs! 🐛

#SoftwareDev #TechHumor #fedi #growth

Hi #FOSS / #SoftwareDev folk!

Please could someone kindly explain what "boilerplate" code is when it comes to software libraries?

Human answers only, please.

Ta x

Anthropic shared an experiment where sixteen Claude agents worked together to build a functional C compiler in Rust. This was not a single model following a script, but a team of autonomous agents using a shared repository to coordinate their work. Over two weeks and 2,000 sessions, they produced 100,000 lines of code capable of compiling a Linux 6.9 kernel.

The significance of this project is not the compiler itself, which still relies on external tools for certain tasks. The real lesson is in the infrastructure required to let AI work in parallel without human oversight. By using Docker containers and simple lock files, the agents managed their own merge conflicts and task distribution. This suggests that the future of development is less about writing individual lines of code and more about designing the environments where agent teams can operate.

🧠 Parallel agents completed a task that usually takes human teams months.
⚡ The $20,000 cost highlights the trade-off between compute and human labor.
🎓 Success depended on automated testing and feedback loops rather than direct instruction.
🔍 The resulting compiler successfully ran Doom and compiled complex open-source projects.

https://arstechnica.com/ai/2026/02/sixteen-claude-ai-agents-working-together-created-a-new-c-compiler/
#AI #Engineering #FutureOfWork #Anthropic #SoftwareDev #Dev #Software