The AI-enabled track
AI Coding & AI Engineering Interviews
Interviews changed. Companies now test how you build with AI — and how you design the systems AI runs on. This track covers both: the engineering judgment for AI-assisted coding rounds, and the system design depth for AI infrastructure roles.
AI System Design Case Studies
The five AI infrastructure designs interviewers ask in 2026 — each with walkthroughs, interviewer scorecards, pressure questions, and free guided practice.
LLM Inference Serving
Continuous batching, KV-cache economics, and the TTFT vs throughput trade.
Case study + guided practice →
RAG Pipeline
Chunking, hybrid retrieval, permission-aware search, and retrieval evals.
Case study + guided practice →
AI Agent Orchestration
Durable runs, idempotent tools, budget governors, approval gates.
Case study + guided practice →
Foundation Model Platform
Training at 10K GPUs: checkpointing math, data pipelines, topology.
Case study + guided practice →
AI Code Assistant
The 200ms completion budget: context engines, prefix caching, cancellation.
Case study + guided practice →
All 21 case studies →New to LLMs?
Learn LLMs: Beginner → Advanced
Six stages from “what is an LLM” to agents and evals. Every stage has a no-code walkthrough for non-developers and a Gemini / Claude / OpenAI code example for developers, plus a six-week plan.
New track
Agent Workflows, Skills & Multi-Agent Systems
How production agents are actually built — workflows vs autonomous loops, reusable skills, multi-agent topologies, context engineering, MCP — with every example implemented for Gemini, Claude and OpenAI side by side.
- Lesson 1 · 14 min readWorkflows vs AgentsMost "agents" in production are workflows: model calls arranged in code-defined paths. Learn the five patterns — chaining, routing, parallelization, orchestrator-workers, evaluator-optimizer — and when to hand control to a real loop.
- Lesson 2 · 12 min readAgent SkillsA skill is a folder of instructions, scripts and references an agent loads only when a task needs it. Learn the SKILL.md format, progressive disclosure, and how to make one skill work across Claude, Gemini and OpenAI.
- Lesson 3 · 16 min readMulti-Agent SystemsSub-agents, supervisors, handoffs and swarms — what each topology buys you, what it costs, and the same research team implemented with Google ADK, the Claude Agent SDK and the OpenAI Agents SDK.
- Lesson 4 · 15 min readContext EngineeringThe context window is the agent’s working memory and its bill. Learn what belongs in it, how to budget it, how prompt caching works on each provider, and how to compact a long-running agent without losing the plot.
- Lesson 5 · 13 min readTools & MCPFunction calling looks different in each SDK but is the same protocol underneath. Learn the request/response shapes side by side, then connect a single MCP server to Gemini, Claude and OpenAI.
- Lesson 6 · 11 min readChoosing a StackA vocabulary map across the three ecosystems — SDKs, agent frameworks, coding harnesses, skill and context conventions — plus a decision framework and the questions interviewers ask about provider choice.
- Lesson 7 · 18 min readPlugins & Agents Hands-OnThree walkthroughs, one weekend: package skills, commands and MCP servers into a Claude Code plugin; ship a ChatGPT app (and Codex plugin) on top of an MCP server; build and deploy a Gemini agent with ADK. Each ends with a shippable artifact.
- Lesson 8 · 15 min readComparing ModelsParameter counts are undisclosed for every frontier model and misleading for the open ones. Learn what each headline benchmark measures, how leaderboards get gamed, how the Gemini, Claude and OpenAI line-ups are tiered, and how to run the only eval that matters — yours.
Coding with AI — what interviewers actually grade
More companies run AI-enabled coding rounds where assistants are allowed. The bar didn't drop — it moved.
Drive the tool, don't be driven
Interviewers watching you code with AI grade one thing above all: who is in charge. Strong candidates decompose the problem first, then direct the assistant at well-scoped pieces — they never paste the whole prompt and pray.
Verify like a reviewer
Every AI-generated block gets the same treatment you'd give a junior's PR: trace the edge cases, question the complexity, run the tests. Accepting wrong code confidently is the fastest fail in an AI-enabled interview.
Know what the machine can't
Requirements, invariants, trade-offs, and system boundaries stay human. The interview signal has shifted from 'can you write a loop' to 'can you specify, verify, and integrate' — which is exactly what our system design track trains.
Narrate your loop
Prompt → inspect → test → refine, out loud. The meta-skill interviewers reward is a tight, articulated iteration loop — the same loop that makes you fast with agents at work.
AI Engineering Guides
Tutorials and interview prep for the agentic stack.
MCP vs A2A: The Two Protocols in Every 2026 Agent Stack, Explained
9 min read · AI Engineering · AI Agents · MCP
AI Agent Memory Architectures: Short-Term, Long-Term, and Checkpointing — What Interviews Actually Test
10 min read · AI Engineering · AI Agents · LangGraph
Google's Agent2Agent (A2A) Protocol, Explained: How AI Agents Talk to Each Other
8 min read · AI Engineering · AI Agents · Agent2Agent
LangGraph Tutorial: Building Stateful, Multi-Step AI Agent Workflows (Free)
9 min read · AI Engineering · LangGraph · Tutorial
LangChain Tutorial: Building Your First AI Agent (Free, Step-by-Step)
9 min read · AI Engineering · LangChain · Tutorial
5 Follow-Up Questions AI System Design Interviewers Always Ask (And How to Answer Them)
7 min read · AI Engineering · System Design · Interview Prep
AI Engineer System Design Interviews in 2026: The Complete Prep Roadmap
8 min read · AI Engineering · System Design · Interview Prep
Write the Agent Loop Yourself: 80 Lines, No Framework
9 min read · AI Engineering · AI Agents · Tutorial
Test yourself under real conditions
Free guided practice on every AI case study — then take a realistic AI mock interview when you're ready to know where you stand.