Last week I decided to turn my 2D game into a 3D one. There is something attractive to me about going vertical in a colony sim. So by lunch I had a plan. That afternoon my AI SDLC pipeline was building the tools that build the tools, two AI subagents were opening pull requests against three different repos, and I had spent an hour and some change making a third assistant argue that the whole pipeline was pointless.<br> More
Category: AI
Inference Costs and Time Travel; 50 hours of work in a day
This post was inspired by a 50-hour day on May 8. By the time I had finished writing it, a more recent peak on May 16 had become the sharper data point. Same arbitrage, tighter compression, but a much higher parallelization. Both days are part of the same story about inference costs and what it is we are actually buying.
On May 16 I shipped 38 hours of work in one day.<br> More
Local AI Agents in LAIRD
I Shipped 12 of Them in a Free Game and Got 2 Downloads
The AAA industry is currently burning billions on cloud compute and server-side LLM APIs. Meanwhile, I just shipped a free game concurrently running 12 AI agents in LAIRD on consumer hardware with zero API latency. It got exactly 2 downloads. I am completely unfazed.
Running 12 Local AI Agents in LAIRD on a Laptop
Ephemeral Working Memory
The Problem with Context Based Memory
Most LLM-powered systems hit a token limit and do the dumbest possible thing: they chop from the front. Take the first N characters, summarize it, discard the rest. Sometimes the logic is slightly smarter: random sample, drop from the middle, oldest-first. But the result is the same. You might silently lose the fact that the player killed the town’s mayor three sessions ago, while keeping five lines of small talk about the weather.<br> More