Last month I spoke at @AllThingsOpen in Raleigh on one of my favorite topics: how AI is transforming code reviews. With tools like Copilot, CodeRabbit, and Qodo, AI can catch bugs, suggest improvements, and even review PRs. So does this replace humans? Not really. AI makes reviews faster, but humans still matter for context, architecture choices, and long term thinking. The future isn’t AI vs humans. It’s AI + humans building better software together. Loved the conversations after the talk. Developers are leaning into this shift with curiosity and responsibility.

Slides: avindrafernando.github.io/th… cc - @cjwhitsitt

Modern AI code review tools need human intervention to get better at their job. Provide them with context, instructions, and learnings. Presented "The Future of Code Reviews" at DevFest KC (@GDGKansasCity). Loved the audience engagement. Slides in post below. #DevFestKC

Want to find out about the future of code reviews? Come on by to Room Blue (352) at DevFest KC at 2:00 pm (central) #DevFestKC

It’s today. See you all soon.

luck is not in your control but, maximizing your chances of getting lucky is 100% in your control

It's not the algo that's broken, it's the notifications. For an app like X notifications are critical.

Happy Friday

If your prompts are wide open and unfiltered, don’t stress about hackers crashing your app. Maybe think about what it can do to your wallet.

Meanwhile @coderabbitai babysitting humans PRs.

How to be a senior engineer in 2025: “You are a senior engineer at XYZ. We’re working on this JIRA together. Always follow AGENTS .md. Make sure any solution aligns with our architecture and best practices. Propose a plan first. Don’t commit without approval.” Know how to talk to AI.

Have you seen this in your circles @tdesseyn ?

In ref to your earlier post today: Hopefully better apps come out of this playbook @Prathkum 😆

Curious if you have started seeing this in your circles, @randomrecruiter @DThompsonDev ?

Vocalizing your thought process is now a critical skill. In the AI era the question isn’t just “can you code?" It’s “can you guide AI to write the right code, and can you explain how you’ll do it?”

If you never question scope or deadlines and just accept them blindly, you’re setting yourself up for pain later. Great engineers call out risk early. Fixed scope + arbitrary deadlines = trouble.

Prediction markets are changing how we think about random events. Now that you can bet on “anything”, and payouts are public, the people/company involved (who the bet is about) could use that prior knowledge to influence the outcome.