Reinforcement learning13 September 2026
Training a Tetrifold Agent with Reinforcement Learning
A technical walkthrough of simulation, linear policies, cross-entropy optimization, and evaluation.
14 min readRead study
06 / Research
HomeLooking closer
at what works.
Longer looks at behavior, trust, and the quiet systems behind products people return to.
Working papers
Questions before answers.
Longer studies for the patterns that are easy to miss in the rush to ship.
Research note2 April 2025
Self-attention is a way of reading.
A visual explanation of how Transformers turn tokens into context—and how that logic runs locally in Pocket AI.
12 min readRead study
↗Blog
Notes from