My Appearance on the Cup O’ Go Podcast

Less than two weeks ago, I had the pleasure of being a guest on Cup O’ Go - one of my favorite Go podcasts, hosted by Shay Nehmad and Jonathan Hall.

Quantum MIME, Eurovision, and lots of MONEY

🤡 Go 1.26.4 and 1.25.11 released🥉 Learning Go, 3rd Edition by Jon Bodner in early release🇩🇪 GopherCon EU, June 15-18 @ Berlin, Germany💰 Reddit: Working with money in Gogithub.com/shopspring/decimal...

https://share.transistor.fm

I have been a fan of the show since its very first episode, and mentioned them in this blog a couple of times. Apparently, they have been reading the blog too, because every now and then, I would see spikes of traffic coming from an episode’s show notes. But the invite to come and talk on the show was a whole new level of recognition. Many, many thanks to you, guys!

Podcasts for Go Developers

For multi-disciplinary creatives who code, photograph, write, and build. Tools, insights, and products for people who don't fit in one box.

https://preslav.me

I went in without a script, and I think it showed - in a good way. Shay and Jon have a gift for making conversation feel effortless, so I just… kept talking. For about an hour.

We covered my early days with Go: how I first encountered the language in late 2018 out of frustration with a painful Python pipeline project. How a CTO at my then-company pointed me toward it, and why my initial reaction was mixed at best.

A decade of Java and Spring had left its mark on me. Unlearning the instinct to reach for abstraction, dependency injection, and five layers of indirection took time. I needed years to understand what Rob Pike meant when he said that “simplicity is complicated.” I’m on the other side of that now - I just like building things, and having Go by my side is the best thing I can do, assuming that moving forward, it will be less of me, and more of a machine writing the code.

And that’s why Go is a perfect match for the age of AI. When a machine writes the code, you want it to be readable - not clever, not layered behind clever patterns, only the author can decode. Go’s lack of magic is its superpower here. An LLM can generate a straightforward Go function and I can review it in seconds, because there are no hidden decorators, no annotation-driven behavior, no framework doing half the work behind my back. What you see is what runs. The same thing that frustrated me in 2019 - the verbosity, the explicitness, the refusal to let me be lazy - is exactly what makes Go the most AI-friendly language I know. The machine writes flat, obvious code. Go was designed for flat, obvious code. It’s a match I didn’t see coming. No one really did.

But Go and AI aside — the moment I’m most proud of in this episode has nothing to do with either. Somewhere in that conversation, I got to pay tribute to a teacher from my middle school - a man who walked into our classroom one day and asked if we wanted to build computer games. We were eleven or twelve. Kids. Of course we said yes. He handed us a thick Delphi book and set us out on a path that, for me at least, never ended. I don’t know where he is today. I hope he’s well, and I hope the message gets to him that I am immensely thankful for that thick Delphi book.

Happy listening!