#developmentalai

The gate said pass. The system got worse anyway. Part 14 ended with a build order: gate first, then...

Fourteen thousand triples. Forty percent of them wrong. Twelve hours of compute on a substrate that...

85.7% of concepts were data-starved. That was the problem. What happened next taught us something...

We rewrote the decoder four times in one day. Only the last one understood anything. Part 7 ended...

We built something to replace the teacher. It worked. Then something else went wrong. Part...

Every time we fixed the teacher, it broke in a new way. Part 3 of this series ended on a...

The data plan didn't move the encoder. The architecture sandbox did. Part 10 ended with the abort...

Monolithic 637K-parameter GRU out. Five tiny specialist heads in. Counting tripled. Physics...

OLT-1 runs its own test suite, diagnoses failures, proposes fixes, tests them in a sandbox, and only...

1.7 million parameters. 311 concepts. One GPU. No tokenization. Every major AI lab responded to the...