Hey,

A lot has happened in the last two months. I have a two month old daughter along with Serena, my two year old. I fully built a new game that I’m in the final stages of refining (as always I keep shifting between loving and hating how it looks). What’s interesting is for this project, I wrote zero lines of code. I used AI for virtually everything -- the visual style, the background music, the sound effects, and all of the content. It was nowhere near as fast as you would think. I started it 2 months ago and have probably spent over 100h on it. It’s a word game for iOS -- if you want to give it a try, let me know! I have a beta running and plan to have it live in the next month or so.

Earlier this year, despite setting a goal to write more, I resigned myself to the fact that I wasn’t really going to write much. I kept having stumbling starts on pieces I wanted to write and never finished them. The good news is I have a lot of pent up writing energy/ideas right now and just need to find a way to channel my energy into finishing some of the many pieces, instead of watching the live action One Piece for a 3rd time with Serena (she has recently become a fan and I can’t say no when she asks to watch it). We have already watched all the Harry Potter movies twice in 2026.

What I’m working on

I mentioned the game already, so I’ll share a bit more about it. One day while picking up Serena from daycare I had the idea to re-do Hangman in a more trendy game style (think Balatro, powerups, etc). Within 24 hours I had it prototyped and a crappy alpha sent to Dylan.

It’s got the 80% same-ness of the final product, with some important changes. Originally I took hangman literally (no hints, only the category) and gave you 7 lives. Either the catalog of words I generated (video games & movies) were way too hard and I suck OR this isn’t that fun. They key change I made was to switch to crossword style clues. From there it really took off. I had some fun trying a more “airport terminal” theme too.

Most of my saved screenshots are of buggy UI so please forgive me.

After over a month of iterating on this, tweaking UI to be more approachable, balancing clue difficulty and fixing bugs (big thanks to all the friends who tested and reported issues), I found it still wasn’t there. The next key change I made was to strip back all the “game” stuff I added and keep it simple, no powerups, no level up system, no challenge words. I left it as a series of clues and surviving up to 30 words. I was starting to sound like an AI there, but I swear I typed those 2 sentences myself.

The evolution over time in small pics

I’m in the final stages of questioning every UI decision. After I finish my two full-AI projects, I’m going tech-Amish and going to try coding something from scratch, sans AI, in a new language.

What I’m reading

My reading pace is super slow this year, but we are getting through books here and there.

Moonbound is Robin Sloan's latest book. He is a super interesting writer - partly of the tech world and the literary one. This novel takes place around 11,000 years in the future and is about a young boy living almost in a Stranger than Fiction world, trying to save humanity. There is a 42,000 dimension space where they communicate with dragons for understanding (because our own 4 dimensions aren’t enough to fully capture the depth of all ideas). There’s this organic technology that is similar to an LLM and I found it very clever. Fun read overall, and if it feels too sci-fi, check out Sourdough also by him.

If you can’t tell, like the rest of the world, I’m on an AI kick. I read The Infinity Machine: Demis Hassabis, DeepMind, and the Quest for Superintelligence, a biography about the founder of DeepMind. Demis is fascinating - seemingly a genius, chess master, previous game designer. He has the most interesting vision, it’s just too bad google’s consumer AI success is so boring. No room for 3rd/4th place. A good intro is one of the two documentaries on YouTube - AlphaGo or The Thinking Game.

You are not in the race against AI slop is one of my favourite pieces of the year - I’ve sent it to many friends. I have an essay inspired by this topic waiting to be finished. This essay argues you don’t need to worry about AI slop if you’re actually committed to great work. Most who are using AI aren’t trying to create something lasting.

I’m coming off a cold and trying to get back into the swing of things while also enjoying the fact that spring has finally come to Calgary.