Learn
The Learn track is sixteen modules, read roughly in order. Each one follows the same structure end to end — an executive summary and mental model up front, an architecture and deep dive in the middle, and failure modes, trade-offs, and interview questions before the hands-on lab. That consistency is deliberate: once you know the shape of one module, you know how to navigate all of them, and how to write the next one.
Required structure
Section titled “Required structure”Every module page carries these sections, in this order. scripts/lint-content-structure.ts fails
CI if one is missing.
- Executive Summary — what this module covers and who needs it, in one paragraph.
- Mental Model — the one idea that makes everything else in the module make sense.
- Architecture — how the pieces fit together, with a diagram.
- Deep Dive — the mechanics: what actually happens and why.
- Implementation — real code, not pseudocode.
- Production Example — how this shows up in a system that takes real traffic.
- Failure Modes — the specific ways this breaks, and how you’d know.
- Trade-offs — what you give up by choosing this approach over the alternatives.
- Security — the attack surface this introduces and how it’s closed.
- Performance — the numbers that matter and how to measure them.
- Scaling — what changes at 10x and 100x load.
- Interview Questions — the questions this material actually answers.
- Hands-on Lab — a link to the companion lab in Build.
- References — primary sources, not blog posts about blog posts.
- Revision History — what changed, and when.
Modules
Section titled “Modules”| # | Module | Status |
|---|---|---|
| 0 | Principal Engineer Mindset | Published |
| 1 | Production Python | Published |
| 2 | Distributed Systems | Published |
| 3 | Networking | Published |
| 4 | AI Infrastructure | Published |
| 5 | Agent Engineering | Published |
| 6 | MCP | Published |
| 7 | LangGraph | Published |
| 8 | RAG | Published |
| 9 | Model Serving | Published |
| 10 | Kubernetes | Published |
| 11 | Cloud | Published |
| 12 | Observability | Published |
| 13 | System Design | Published |
| 14 | Leadership | Published |
| 15 | Agent Identity and Access | Published |
All sixteen modules are published. See the Roadmap for what is being built next.
Decision models
Section titled “Decision models”Decision models — small, fast models that answer a bounded question with a probability instead of generating text — are taught inside existing modules rather than in a module of their own. Module 4 places a decision tier between rules and an LLM, and Module 5 covers the failure mode of treating a confident answer as a correct one. The vendor numbers are in the Decision Models lookup, and the measurement is in the Decision Tier Calibration lab.