Netflix AI LLM Engineer Interview Questions

Self-hosting Llama 3 is a infrastructure-heavy question for ML platform and AI engineer roles at Meta-adjacent companies. Expect deep dives on quantization, vLLM, GPU sizing, and the TCO math that determines build vs buy decisions.

Hard Llama/inference question on continuous batching and GPU efficiency.

**Metrics** Requests queued, p95 latency, GPU memory headroom.

**Training** Rank selection, learning rate, epochs; validation loss + task evals.

**vLLM** High throughput continuous batching; popular for multi-tenant APIs.

**Eval design** Expert-labeled scenarios; citation accuracy checks; refusal appropriateness.

**Structure** Eval templates + data JSONL + scoring functions; CI integration via API.

OpenAI's model lineup changes fast. GPT-4 vs GPT-4o is a model selection question that tests whether you understand latency, cost, multimodal capabilities, and when reasoning depth matters. Critical for any role touching OpenAI APIs in production.

Constitutional AI is Anthropic's differentiator and a must-know for Claude-focused interviews. Go beyond the marketing — explain the self-critique training loop, how CAI compares to RLHF, and practical safety implications for production deployments.

Google's Gemini 1.5 Pro long-context window opens use cases impossible with standard LLMs — whole-codebase analysis, multi-hour video, massive document review. Interviewers test whether you understand real limitations behind the 1M token marketing number.

Claude Code is Anthropic's CLI coding agent designed for repository-level work: reading the codebase, editing multiple files, running tests/builds, and iterating until a task succeeds. Unlike autocomplete-first copilots that live inside the editor, Claude Code operates as an agent loop — observe → plan → act (edit/run tools) → verify.

A production Claude Code rollout is an agent platform problem, not a 'give everyone a CLI' problem.

Codex CLI represents OpenAI's push into agentic coding outside the editor — similar category to Claude Code and other CLI agents. The core loop is: understand task → explore repo → edit → run commands → iterate.

Cost optimization for coding agents is context engineering + runtime policy.

Medium Gemini/Vertex question on context caching vs RAG trade-offs.

Medium Claude cost question on prompt caching breakpoints and hit rates.

Medium GPT API question on structured outputs reliability vs JSON mode.

**Dimensions** Quality, cost curve, latency, compliance, ops burden, roadmap velocity.

**Analysis** Rank features by volume, quality bar, data sensitivity benefiting on-prem.

**Key clauses** 700M MAU threshold triggers extra terms; AUP prohibits harmful uses.

**Usage** Separate small model for moderation latency.

**Constraints** No phone-home telemetry; offline license compliance documentation.

**Training** Refusal datasets, toxicity filters on training data.

**GPTQ/AWQ** GPU inference friendly; integrate with vLLM/TGI depending on support.