Testing certain type of games with LLMs has never been easier, if there's randomness -> use seeds, then you can define some rules for each strategy and automate the runs via scripts for easy before/after comparisons when i.e. tweaking costs in an incremental game. Also game time:
41
Unpopular opinion time, but anybody that actually codes and works in a development industry, and actually read this article, would agree this doesn’t matter at all.
Please stop punishing developers for using code assist tools… Punish them for generating music and art.
Big Walk used AI in 'the final stretch' of development, kicking off intense awards debate:
aftermath.site/big-walk-ai-i…
Readers added context they thought people might want to know
House House used LLMs sparingly near the end of Big Walk's development, mainly to fix bugs in third-party code. They estimate well under 0.1% of the game's code was affected, and they didn't use generative AI elsewhere.
bigwalk.game/faq/#ai
pcgamer.com/games/puzzle/b…
Final form: Compose multiplatform for desktop using the CLI in bend.
Replying to @aballano
Got it working now, no bugs, new features like manual time adjust are also easy to encode following Laws 😍
1
87
Running opencode from another laptop I was wondering how to allow certain sudo commands without sharing my pass...
Custom command queue + ntfy + connectbot = I get notifications on my phone with the exact sudo command so tapping and typing my pass over ssh approves it 🤯
83
Alberto Ballano retweeted
For all the parents out there who might have missed it. We just rolled out one of our all time most requested features.
There is finally a toggle in your settings to automatically “Exclude all kids music from my taste profile”
Now go reclaim your account!🎧🎶
I was about to cleanup/optimize my TVs and just discover this beauty
github.com/Gathin23/cleanyou…
2
161
Alberto Ballano retweeted
Replying to @catpublic10
Without focusing too much on it being the biggest copyright-laundry scheme ever, and also not speaking as a teacher & how AI harms the learning process by removing obstacles that are vital for deep/healthy learning, the main argument is mostly related to control and technological dependence.
A technology can make you more efficient while simultaneously making you less capable of functioning without it.
With AI coding, the concern is that if you outsource the planning, debugging, architecture, and implementation often enough, you may gradually stop developing those abilities yourself. Eventually the tool goes from "assistant" to "something I cannot work without."
The burning question here is... is the technology increasing your capabilities, or replacing them?
Most people were already slop coders and did not care about the underlying understanding of their code before 2025 anyway, but we still need kids that know every inner detail of how technology works to make contributions to real breakthroughs and proper creative innovation in the future.
Ok, reviving a 2014 MacBook Pro on late 2026 using Linux definitely wasn't in my bingo card, but damn does it feel nice now
2
91
Alberto Ballano retweeted
Bend 2 is here!
It is a new programming language that blocks AI mistakes via *proof checking* - the same technique big AI labs used to solve open math problems, like Navier-Stokes.
It is also very fast, and runs on GPUs.
Watch the video. Link in the comments.
RELEASE DAY
After almost 10 years of hard work, tireless research, and a dive deep into the kernels of computer science, I finally realized a dream: running a high-level language on GPUs. And I'm giving it to the world!
Bend compiles modern programming features, including:
- Lambdas with full closure support
- Unrestricted recursion and loops
- Fast object allocations of all kinds
- Folds, ADTs, continuations and much more
To HVM2, a new runtime capable of spreading that workload across 1000's of cores, in a thread-safe, low-overhead fashion. As a result, we finally have a true high-level language that runs natively on GPUs!
Here's a quick demo:
























