Quick references for your next loop.

01 · Cheatsheet

LLM Fundamentals

Core concepts behind modern large language models, from transformer architecture to inference. Interviewers test whether you can explain what happens inside the model at a mechanistic level.

Transformers Attention Context windows Pretraining Tokenization Inference

02 · Cheatsheet

Prompt Engineering

Techniques for reliably controlling LLM behavior through prompt design. Interviewers look for systematic thinking about instruction clarity, robustness, and measurable evaluation.

Instruction design Few-shot prompting Guardrails Output schemas Prompt evals Tool prompts

03 · Cheatsheet

RAG

Patterns for grounding LLM responses in retrieved documents. Interviewers probe your ability to identify and fix quality bottlenecks across the retrieval and generation pipeline.

Chunking Re-ranking Hybrid search Freshness Citations Latency budgets

04 · Cheatsheet

AI Agents

Design patterns for LLM-driven agents that plan, use tools, and complete multi-step tasks. Interviewers focus on reliability, failure modes, and evaluation of open-ended loops.

Planning Tool use Memory Multi-step loops Failure recovery Agent evals

05 · Cheatsheet

AI Agent Reliability

Production patterns for building agents that survive the real world — idempotency, audit trails, approval gates, and clean architecture boundaries. Interviewers test whether you know how agents fail and how to design around those failures.

Agent Reliability Tool Use & Permissions Agent Architecture

06 · Cheatsheet

Fine-Tuning

Techniques for adapting pretrained models to new tasks or domains. Interviewers test your ability to choose the right adaptation strategy and diagnose training problems.

SFT LoRA QLoRA Dataset quality Hyperparameters Overfitting

07 · Cheatsheet

Vector DBs

How to store, index, and retrieve dense vector representations at scale. Interviewers test your understanding of ANN tradeoffs, embedding model selection, and operational concerns.

Embedding choice ANN indexes Metadata filters Recall Dimensionality Updates

08 · Cheatsheet

AI System Design

End-to-end architecture of AI-powered systems at production scale. Interviewers expect you to reason about serving, caching, multi-tenancy, fallbacks, and observability.

Serving RAG topology Caching Multi-tenant Fallbacks Observability

09 · Cheatsheet

LLMOps

Operational practices for shipping and maintaining LLM-powered systems reliably. Interviewers look for experience with rollout safety, cost control, and regression prevention.

Rollouts Tracing Cost control Prompt versioning Safety checks Regression suites

10 · Cheatsheet

Evaluation

Rigorous methods for measuring LLM system quality. Interviewers probe your ability to design reliable evals, choose appropriate metrics, and avoid evaluation pitfalls.

Golden sets LLM-as-judge A/B tests Offline metrics Error taxonomy Sample size

11 · Cheatsheet

Safety & Ethics

Principles and practices for building AI systems that are safe, fair, and trustworthy. Interviewers test your ability to identify risks and implement mitigations in real products.

Prompt injection Refusals PII Policy design Bias checks Red teaming

12 · Cheatsheet

Multimodal AI

Systems that process and generate multiple data modalities — images, audio, and text. Interviewers focus on fusion strategies, modality-specific failure modes, and latency.

Vision inputs OCR Speech Fusion Latency Grounding

13 · Cheatsheet

AI Infrastructure

Hardware, orchestration, and cost strategies for running AI workloads at scale. Interviewers test your ability to design systems that remain performant and cost-efficient under load.

GPU scheduling Batching Rate limits Autoscaling Queues Cost per request

14 · Cheatsheet

Coding

Engineering fundamentals for building AI systems in production Python. Interviewers often give live coding tasks covering async I/O, streaming, type safety, and API integration.

Async I/O Streaming Typing Testing Profiling APIs

15 · Cheatsheet

Behavioral

Soft-skill and leadership questions interviewers use to assess culture fit and engineering maturity. Prepare specific stories using the STAR framework for each focus area.

Ownership Tradeoffs Conflict Prioritization Ambiguity Impact