Is What I Do Still Going to Matter?

I’ll be honest with you.

When I started seeing what AI could do to software development, my first reaction wasn’t excitement.

It was anxiety.

I’ve spent over a decade building and leading .NET/Azure development teams. That experience is my livelihood. And suddenly the tools I’d built a career around were being automated, summarized, and second-guessed by a chatbot.

I know I’m not alone in that feeling. I’ve talked to enough engineering managers and senior developers to know a lot of us are quietly asking the same question:

“Is what I do still going to matter in five years?”

I sat with that question for a while. And then I decided to stop waiting for the answer and go find it myself.

So I built something.

DevTeam AI Assistant

DevTeam AI Assistant is a .NET 8 console app that uses the Anthropic Claude API to do three things most dev teams do manually today:

Analyze sprint retrospectives for patterns, blockers, and action items
Review code diffs for readability issues, bugs, and architectural concerns
Score and prioritize technical debt by impact vs. effort

No third-party AI SDKs. Just C#, HttpClient, and a lot of time spent learning how to structure a prompt that actually produces useful output.

What building it taught me

AI doesn’t make experienced engineers irrelevant. It makes inexperienced judgment more dangerous.

The retro analyzer surfaces patterns — but you still need someone who understands team dynamics to act on them. The code reviewer flags issues — but you still need someone who knows the system to evaluate the tradeoffs.

The engineers and leaders who will thrive aren’t the ones who ignore AI or the ones who blindly trust it. They are the ones who understand it well enough to know when to use it and when to push back.

What this means for IdeaRoost

That’s what I’m building IdeaRoost around — and I’m committed to practicing what I preach. Claude Code is now a standard part of my development process. Every engagement, every build, every project. Not as a shortcut, but as a force multiplier for the judgment and experience I bring to the work.

I built this entire project using Claude Code as my development assistant — the first time I’d used an AI tool end-to-end on a real build, not just for a quick snippet, but for architecture decisions, debugging, and refactoring.

I have a lot of thoughts about what worked, what didn’t, and what it means for how development teams will actually operate in the near future.