← Interview Preparation

AI Engineering Reference

How to work with AI as a professional LLM user in mid-2026 — what models are and where they fail, prompting and context engineering, agentic IDE workflows (Cursor), RAG and internal knowledge bases, MCP tools and automations, multi-agent patterns, model selection, and team governance. Not vibe coding; not training transformers. Anchor sources: Anthropic, OpenAI, Google Gemini, Cursor, and modelcontextprotocol.io.

Time budget: ≈34h

How to use this reference

  • Work through topics top to bottom — MCP and agents assume you understand tool use; RAG assumes you understand embeddings and hallucination limits from Topic 1.
  • Every subtopic teaches what the model is doing so you can verify outputs, set guardrails, and choose the right integration pattern — not copy-paste prompts blindly.
  • Vendor docs (Anthropic, OpenAI, Cursor) change fast — our overviews point to durable concepts; follow the linked official docs for API details.
  • Optional tasks include local MCP experiments and workflow design drills — skip what you already do daily at work.
  • Building production AI products at scale (serving, evals, rollouts) is a separate track — continue to AI Systems Reference after this one.

The Reference

  1. 1

    Transformers, tokens, and inference — what the model actually does before you prompt it.

    1. 1.1How LLMs Work (Inference, Not Training)2/530m
    2. 1.2Capabilities, Limitations & Hallucinations3/530m
  2. 2

    Prompts are API calls — context is the payload. Design both like you design interfaces.

    1. 2.1Prompt Engineering for Code3/530m
    2. 2.2Context Engineering & Specs4/535m
  3. 3

    IDE agents, Cursor workflows, and the daily loop — where AI earns its place in your toolchain.

    1. 3.1IDE Agents & Cursor Workflows3/530m
    2. 3.2Review, Testing & Refactoring with AI3/530m
  4. 4

    Ground the model in your docs and code — retrieval beats bigger context windows.

    1. 4.1RAG Fundamentals & Chunking3/530m
    2. 4.2Internal Docs & Codebase Search4/535m
  5. 5

    Give the model hands — tool use, MCP servers, and CI integrations that turn chat into action.

    1. 5.1MCP & Tool Use Protocol4/535m
    2. 5.2Automations, CI & Plugins4/535m
  6. 6

    Single agents hit ceilings — orchestration, handoffs, and evals for production agent systems.

    1. 6.1Agent Architectures & Patterns4/535m
    2. 6.2Production Agents & Evals5/535m
  7. 7

    Route the right model to the right task — and know when fine-tuning beats RAG beats prompting.

    1. 7.1Choosing & Routing Models3/530m
    2. 7.2Fine-Tuning vs RAG vs Prompts4/535m
  8. 8

    Scale AI adoption across a team — rules, skills, security, IP, and policies that outlast any model version.

    1. 8.1Rules, Skills & Project Context3/530m
    2. 8.2Security, IP & Team Adoption4/535m