MLOps

Pick a topic and flip the cards.

5 decks · 250 cards

These cards are original study material. They are not questions from a named employer or a live hiring loop.

• AI agents basics

  The runtime facts of an agent harness: tools, limits, state, traces, and rollouts.

  50 cards

• Context engineering basics

  Token budgets, prompt caches, retrieval indexes, and the KV cache that serves a context.

  50 cards

• Model monitoring

  Drift, delayed labels, shadow and canary releases, and alerts worth paging on.

  50 cards

• Training pipelines

  DAGs, data contracts, snapshots, artifacts, registries, and what gets promoted.

  50 cards

• Transformer basics

  The memory, numerics, parallelism, and artifacts that a transformer train or serve stack depends on.

  50 cards