← Speculative & Constrained Decoding / 20

hardNewBasetenSnowflakeAmazon

Implement a JSON-schema token mask: walk the vocabulary through the automaton, pack it into int32 words, apply it to logits.

The coding round behind structured output: a character automaton for a small schema, a vocabulary trie that prunes dead prefixes, the int32 bitmask layout engines ship to the GPU, and the four edge cases that produce invalid JSON or a crash when missed.

Updated Oct 2026 · Worked answers on inference mechanisms, with calculations, implementation details, and follow-up questions.

10 answers per topic instead of 5, plus saved progress and bookmarks · costs nothingor unlock all 213 remaining answers · $25

Every worked answer and derivation in all 12 topics, every concept page, the full practice-test bank, and every code drill’s edge-case tests and reference solution.

6 months · One payment · No auto-renewal

UP NEXT ON YOUR JOURNEY

Next in this trackA diffusion LLM emits 15 to 20 tokens per forward. Is it 15x faster than autoregressive decode? Derive where it wins.Next in this trackAn RL rollout engine speculates with an EAGLE head trained on the step-0 policy, and rollouts are no faster now. Why?Popular right nowWhy is batch-1 decode on a 70B model bandwidth-bound, and what is its latency floor on two H100s?

EDITOR'S NOTE

The bug interviewers wait for is checking only a token's first character. The second is the special token with empty text that every walk accepts.

DISCUSSION · 0

Nobody has commented on this answer. Add a correction or another way to work it.