AI/ML Engineer

Pick a topic and flip the cards.

10 decks · 500 cards

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

• AI agents basics

  The agent loop, tools, the harness around the model, and how a run stops.

  50 cards

• AI/ML system design

  The parts of a system that serves a model, from the request path to retrieval, caching, and capacity.

  50 cards

• Context engineering basics

  Prompts, retrieval, and the other pieces that fill a model's context window.

  50 cards

• Deep learning architectures

  Building blocks of neural networks, convolutional and recurrent designs, and the models that followed.

  50 cards

• Evaluation and data

  Splits, labels, leakage, slices, and the difference between offline and online scores.

  50 cards

• LLM fine-tuning and hallucination

  Ways to adapt a language model, how preference tuning works, and what hallucination is and how to reduce it.

  50 cards

• ML fundamentals

  Bias and variance, regularization, validation, and the metrics used to judge a model.

  50 cards

• Production readiness

  Reproducibility, release checks, rollout, serving, and monitoring for a model in production.

  50 cards

• Statistics and A/B testing

  Distributions, estimates, hypothesis tests, and the parts of an online experiment.

  50 cards

• Transformer basics

  Tokens, attention, positions, masks, and the parts of one transformer layer.

  50 cards