interview-prep

A reference and drill surface for AI PM interview rounds. Distilled from Aakash Gupta's coaching rubric (calibrated against 200+ candidates / 30+ AI PM offers at OpenAI, Anthropic, Google DeepMind, Meta AI, Amazon AGI).

Live content:

• The 7 interview rounds + 8 behavioral dimensions + 3 Laws
• The DASME framework — 4-layer system-design diagram and 7 anti-patterns
• The model-selection table (XGBoost vs LLM vs rules) — the single most testable piece of knowledge
• The SIGNAL metric cascade (model → UX → business)
• 64 system-design questions + 8 product-sense bonuses, filterable, with a 45-minute drill timer that surfaces the DASME phase you should be on
• Company playbooks for OpenAI, Anthropic, Google DeepMind, Meta AI, Amazon AGI, Netflix, Apple, Nvidia
• Four paired calibration answers — 4/10 next to 9/10 — with annotations

Stack

• Next.js 16 + Turbopack
• React 19
• Tailwind CSS v4 (OKLCH theme tokens)
• TypeScript
• All content is statically rendered — no backend, no database

Design

The visual system follows the principles in ryanthedev/design-for-ai and pbakaus/impeccable:

• OKLCH throughout, with tinted neutrals (warm cream paper / warm dark ink)
• One intentional accent — burnt sienna — instead of a default blue, teal, or purple gradient
• Geist Sans for UI, Fraunces for display (variable, with SOFT and WONK axes engaged on the italic for the editorial feel)
• Modular type scale, generous spacing, hairline rules
• Calm easing curves, no bouncy or elastic motion
• Honest dark mode driven by prefers-color-scheme

Run

npm install
npm run dev    # http://localhost:3000

npm run build
npm start

Layout

src/
├── app/
│   ├── page.tsx                   # Home — hero, 3 Laws, 7 rounds, 5 shifts
│   ├── questions/                 # Question bank + drill timer
│   ├── frameworks/                # DASME, SIGNAL, models, anti-patterns, safety, vibe
│   ├── companies/                 # 8 company playbooks
│   └── calibrations/              # 4 paired answers (4/10 vs 9/10)
├── components/
│   ├── site-nav.tsx
│   └── site-footer.tsx
└── content/
    ├── laws.ts                    # 3 Laws, 8 dimensions, 7 rounds, 5 shifts
    ├── questions.ts               # 72 questions across 7 categories
    ├── companies.ts               # company playbooks
    ├── calibrations.ts            # paired calibration answers
    └── frameworks.ts              # DASME, model table, SIGNAL, safety, vibe

Companion CLI

This UI is a study and reference surface. The companion Python CLI at ai-pm-interview-prep/ (separate folder) does the personalised work — pulling stories from a wiki, ranking them by JD keywords, and assembling a context pack to paste into a Claude chat for grading by the ai-pm-interview-coach skill.

License

MIT for the code. Content is adapted from publicly available coaching material; respect the source.