Home
Guide
A self-paced engineering degree: systems, infrastructure, machine learning, AI and inference engineering, plus the product and startup knowledge to use them well.
Program map
Foundations of the craft
Build the single-machine and single-hop mental model (CPU, memory, OS, networking, storage engines) alongside the day-to-day craft of a professional engineer: levels, ownership, maintainable code, testing, code review and debugging. Everything later assumes both, and the craft half pays off at work immediately.
Backend systems and production
The core of the SDE2 jump: design distributed systems and run them. System design and infrastructure are taken together because each explains the other: replication and consensus show up again as Kubernetes and etcd, and SLOs give 'availability' a number. Add the career pages on owning projects (design docs, planning, production ownership) here.
Machine learning
Learn how models actually work, from the math through backpropagation, PyTorch and the transformer to pretraining, scaling laws and post-training. Doing this before AI engineering means prompts, RAG and fine-tuning become engineering decisions rather than recipes, and it is required for inference engineering.
AI engineering and product
Build products on LLMs: prompting, context engineering, retrieval, agents, MCP, evals, safety and production operations. Product is paired here because AI features succeed or fail on problem selection, metrics and UX as much as on models; the product course also teaches the experimentation and metrics needed to judge them.
Inference engineering
The specialist track: GPUs, the roofline model, the arithmetic of prefill, decode and the KV cache, then batching, PagedAttention, FlashAttention, quantization, speculative decoding, parallelism, disaggregation and serving economics. It draws on foundations (memory hierarchy), infra (Kubernetes, GPU infrastructure, autoscaling) and ML (transformers), so it comes last among the technical courses.
Business, funding and the path to SDE3
Understand how the companies you build for make money and get funded: business models, unit economics, AI margins, SaaS metrics, go-to-market, then SAFEs, term sheets, cap tables and employee equity. Finish career pages on growth and interviewing, which now have the technical and business depth behind them. If you are interviewing sooner, pull the career interviewing pages and the system design case studies earlier.
Want everything on one page? Read the full syllabus.
Topics
Computer systems foundations
What actually happens between your code, the CPU, the OS, the network and the disk.
System design
How large systems stay fast, available and correct, from building blocks to full case studies.
Infrastructure, cloud and SRE
Running software in production: Linux, cloud, Kubernetes, delivery, observability, reliability, security and cost.
Machine learning
From the math you need to training transformers: classical ML, deep learning, LLM pretraining and post-training.
AI engineering
Building reliable products on large language models: prompting, context, retrieval, agents, evals and production.
Inference engineering
Serving models fast and cheaply on GPUs: hardware, the math of LLM inference, optimizations, serving systems and economics.
Engineering craft and career
The skills that get you from junior to SDE2 and on track for SDE3: craft, ownership, communication and interviewing.
Product
Product thinking for engineers: finding problems worth solving, measuring what matters, experimenting, growing and pricing.
Startup economics and business
How startups make money: markets, business models, unit economics, financial statements, SaaS metrics and go-to-market.
Venture funding and company building
How venture capital works and how to raise it: SAFEs, priced rounds, term sheets, cap tables, employee equity and exits.
Start here
Back-of-the-envelope estimates
Turn users and features into QPS, storage and bandwidth numbers in two minutes, and use them to decide what the design actually needs.
Caching
Where to put a cache, how to read and write through it, how to size it from hit-rate math, and how to keep it from serving stale data or melting down.
Case study: URL shortener
The classic warm-up: ID generation, read-heavy caching, 301 vs 302 redirects, and analytics without slowing the hot path.
How transformers work (practitioner view)
Just enough transformer internals to predict model behavior — attention, context, and why the KV cache shapes pricing.
Retrieval-augmented generation
The core RAG pipeline: ingest, chunk, embed, retrieve, generate with citations.
Tool use and function calling
Letting models call your code: tool schemas, the tool loop, and designing tools models use well.