Articles by Grizzly Peak Software

I turned Kodiak, my open 1B decision model, into a local judge API: it picks the better of two answers in about 50 ms, accepts only the verdicts it's sure of, and catches its own position bias by judging every pair both ways.

I made Kodiak, my open-weights decision model, and Jev, TypeSafe AI's closed one, play Zork by choosing from the game's valid commands: same harness, same seeds. In about 20,000 moves, not one was invalid. Jev played better, then walked into a grue while its own danger answer said "yes" at 0.95. Here's what that, a no-model baseline and a 35x question-wording effect taught me about System 1 / System 2 cascades.

Kodiak is an open-weights decision model: give it a document and typed questions, and it answers all of them in one pass, in 8 ms, with calibrated confidence and an honest "I can't tell." I built it on my DGX Spark with Claude Code in three days. Here's how it works, what it costs, a 3 a.m. plot twist, and where it stands against bigger models.

I got Eon Systems' whole-brain fruit fly emulation running on an NVIDIA DGX Spark: 138,639 neurons wired from a real fly's connectome. It reproduced the paper's sugar-to-feeding result, needed a CUDA 13 fix to build the faithful GPU path, and ended up driving a simulated body that walks and turns on its own descending-neuron signals.

A late-night dispatch from an Alaska cabin covering AI-powered productivity, marketing lessons learned the hard way, and building in the quiet.