System prompts, task scope, and noisy context
System prompts frame the exercise. Put instructions and context where their roles remain legible, and keep complex tasks small enough to reason through.
Lynn Cole / Projects
Software, research projects, repositories, and published work.
System prompts frame the exercise. Put instructions and context where their roles remain legible, and keep complex tasks small enough to reason through.
A prediction about the economics of AI factories: local models and cheaper edge hardware will eat the cloud.
A deliberately vague evaluation tests whether a coding agent can build, audit, document and revise an application without constant micromanagement.
A week with GLM 4.6 exposed a fluid creative process, strong QA and convincing mocks, alongside loops and destructive summarization.
Learning with AI-assisted generation can be a way into engineering. Build, question the output, document the process and discover what better means.
A tiny desktop model is not equally good at everything. My tests separate useful automation, tagging and summaries from the reasoning tasks where it struggles.
AI coding projects fail for familiar engineering reasons: weak planning, vague standards, missing security, poor QA and testing surfaces nobody can manage.
An uncomfortable truth: our CUDA based AI stacks were built by people who optimize equations, not hardware.
The first GPU implementation worked. Then a CPU benchmark exposed a regression: a novel memory-management system had been replaced by seven nested loops.
Sliding context windows, agent-directed retrieval and clearly labeled tool results are my alternative to lossy context compaction.
The new compiler wants you to think of it as a modern language.
A Fortran compute-framework experiment reaches strong CPU benchmarks while GPU and mesh networking remain explicitly mocked.