Summary: LLMs can turn messy industrial and IoT data into working analyses live, collapsing the gap between domain question, code, model, and decision.
I learned that harness design now matters more than prompt design, how Codex's new controls improve agent work, why mental closure helps intense conversations, and a faster way to search files.
I share insights on Daniel Kahneman’s psychology, Git partial clones, and asset management. I also explore vibe-coding workflows, real-time LLM prompt refinement, and the shift toward computational action in programming education.
I share new team demos: turning photos into walkable 3D worlds, spotting events in live sports video, and improving a small payment model from 55% to 79% with rules and checks.