LLM / Foundation-Model Interview Prep
Concentrated reps on the topics that actually get asked.
| File | Purpose |
|---|---|
| 01-concepts-cheatsheet.md | One-page answers to the Top 20 Questions from the master README |
| 02-llm-coding-questions.py | Implement-from-scratch challenges (attention, KV-cache, BPE, top-p, beam search) |
| 03-systems-questions.md | Performance, parallelism, memory, profiling deep-dives |
| 04-system-design-walkthroughs.md | Cross-references the system-design/ folder with practice prompts |
| 05-research-engineering-questions.md | Pretraining-engineer specific: numerical stability, scaling laws, debugging |
| 06-behavioral-questions.md | STAR-format frameworks for AI-org behavioral rounds |
Recommended Schedule (4 weeks before interviews)
| Week | Focus |
|---|---|
| 1 | 01 + 02 — make sure you can write attention, KV-cache, BPE on a whiteboard |
| 2 | 03 + 04 — performance + 2 system-design walkthroughs cold |
| 3 | 05 + remaining system-design — practice "I don't know but here's how I'd find out" |
| 4 | 06 + mock interviews — prepare 4 stories covering: ambiguity, cross-team, failure, impact |