I stopped treating Clap like a dataset and treated it like an operator system
Clap evolved from dataset plumbing into a private-first operator system with simulation-first execution and local training paths.
Mar 311 min read
QERS got real when I focused on gap metrics, not simulator hype
QERS moved from architecture pitch mode to measurable profile-based simulator evaluation with practical gap metrics.
Mar 311 min read
The NCAA model improved after I split the problem into seed zones
A practical update on tiered seed-band modeling, routing logic, and assignment optimization for NCAA seed prediction.
Mar 311 min read
Weeknotes: 2026 W14
A weekly live update covering technical progress, project movement, friction, and next focus.
Mar 311 min read
LOL News and the problem with novelty-only apps
LOL News started as a playful React Native app idea. The useful work was in feed reliability, voice output behavior, and content filtering choices.
Mar 311 min read
Convo AI was fun until progression logic got real
Convo AI started as a gamified learning prototype. The hard part became progression logic, state consistency, and writing tasks that do not feel fake.
Mar 311 min read
Building an ethics tool inside the tab people already use
AI Ethics Coach is a Chrome MV3 extension that reviews selected text and returns practical flags without pretending to replace policy or judgment.
Mar 311 min read
_edited_edited.jpg)
