Ultimate AI Engineer Flashcards

In LoRA (Low-Rank Adaptation) fine-tuning, what is decomposed into low-rank matrices?

Answer: The weight update matrices ΔW

LoRA freezes original weights and approximates ΔW = A·B where A and B are low-rank, drastically reducing trainable parameters.

Which of the following is an example of few-shot prompting?

Answer: Including 3 example input-output pairs in the prompt before the actual query

Few-shot prompting embeds a small number of demonstration examples directly in the prompt to guide the model's output format and style.

A production LLM application has a P99 latency of 8 seconds. Which optimization most directly addresses time-to-first-token?

Answer: Applying KV-cache to the prefill step

Caching the KV states during prefill avoids recomputing them for the prompt, directly reducing time-to-first-token latency.

What problem does RLHF (Reinforcement Learning from Human Feedback) primarily solve?

Answer: Aligning model outputs with human preferences and values

RLHF trains a reward model on human preference data and uses RL to fine-tune the LLM to maximize that reward, improving alignment.

Which evaluation approach detects when a model performs well on a benchmark due to memorizing test data rather than generalizing?

Answer: Contamination analysis checking if test data appears in pretraining

Contamination analysis searches the pretraining corpus for n-gram overlaps with benchmark test sets to detect data leakage.

In a multi-agent AI system, what is the role of an 'orchestrator' agent?

Answer: Decomposing tasks and routing sub-tasks to specialized agents

An orchestrator decomposes complex goals into sub-tasks and coordinates specialized sub-agents to execute them.

What does 'context window' limit in a large language model?

Answer: The total tokens (prompt + completion) the model can process in one forward pass

The context window defines the maximum sequence length the model can attend to in a single inference call, encompassing both input and output tokens.