Overview
This book collects my share-safe technical notes for the AI Singapore AIAP-style portfolio work: methods, how I would ship it, systems-scale interview angles, and a gauntlet of questions answered against the actual codebase—not the assessment brief.
After sign-in: flashcards and a timed mock on the admin quiz; optional browser voice practice (TTS + speech-to-text warm-up) at /admin/aiap/interview; timeline, articles, teleprompter, and bookshelf on /admin/aiap.
Chapters
- Methods — Tabular pipeline, features, models, leakage experiment, diagnostics. Dotted terms open repo code excerpts.
- Production deployment — Train vs serve, artifacts, preprocessing contract, inference.
- Interview — systems — 100× batch + streaming, cost, CDC, incremental learning.
- April 25 gauntlet — Projected questions answered against the real take-home.