I care a lot about the tools I use. Especially when they aren’t free.

Here’s what I’m paying for in 2023–24 to improve my performance and productivity as a software engineer.

Please note: None of the links in this article are affiliate links.

GitHub Copilot: an AI pair programmer

When I’m writing code, Copilot works in the background by reading what I’ve written and quietly suggesting what I might want to write next.

Last year I said Copilot has improved my productivity by at least 30%. The real figure might be closer to 50%. I spend considerably less effort on the mundane and boilerplate, and I feel much more satisfied with the engaging bits of software engineering.

My biggest productivity improvements have come from:

• Writing test cases. Most of the time I write the description for the test case and Copilot fills out everything else.
• Small things I would have to look up. e.g. Instead of searching for the correct RegEx to parse a string, I write a comment explaining what I want the RegEx to do and Copilot writes the RegEx for me.
• Boilerplate functions. For 90% of boilerplate functions (e.g. snakeCaseToCamel or loggingMiddleware) I write the function name and Copilot writes the function.

Learn more about GitHub Copilot →

Kagi: a better search engine than Google

I measure the effectiveness of searches by how long it takes me to find what I was actually looking for. By that measure, Google has been steadily getting worse.

When I search for something on Kagi, the correct result is in the first 2 links 95% of the time. It’s in the top 5 links 99% of the time. That just doesn’t happen with Google, Bing, etc.

The consistently great results page is further boosted by the search personalization I control. I’ve told Kagi that any results from Stack Overflow or Medium should be weighted higher, as well as blocked other sites I don’t care to see…