The knowledge library

Build the mental models behind modern software.

Follow a guided path through connected systems, or explore one essential concept at a time. The goal is to help you understand, build, and debug software independently in the AI era.

01 · Start here

See the whole request before studying the pieces.

A guided sequence for understanding how the layers of a web application cooperate—and where they fail.

Guided path Start here

How a Web Application Actually Works

Follow a request from the browser through DNS, HTTP, the backend, the database, deployment, and back to the user.

10 published lessonsAbout 3 hr 20 min

Open the path overview

02 · Foundations

A working map of the software stack.

Categories organize the territory. Each one begins with a small set of concepts that make the next layer easier to reason about.

  1. How Software Works

    Build a systems-level picture of how programs use machines, communicate, and persist information.

    • Client and server
    • Processes and threads
    • Memory, CPU, and storage
    Explore How Software Works
  2. Frontend

    Understand how browsers turn code and data into interactive interfaces.

    • How Browsers Turn HTML and CSS Into Pixels
    • The DOM, Events, and User Interaction
    • The JavaScript Runtime, Event Loop, and Async Code
    Follow the frontend learning path Explore Frontend
  3. Backend

    Trace what happens after a request reaches the application server.

    • Request lifecycle
    • APIs and webhooks
    • Authentication and authorization
    Explore Backend
  4. Databases

    Learn how software organizes, finds, protects, and changes durable data.

    • Data modeling
    • Indexes
    • Transactions
    Explore Databases
  5. Cloud and Deployment

    See how an application moves from source code to reliable running infrastructure.

    • Virtual machines and containers
    • DNS and reverse proxies
    • Regions, Availability Zones, and CDNs
    Explore Cloud and Deployment
  6. System Design

    Reason about the tradeoffs that keep software responsive, resilient, and understandable.

    • System Design Starts With Requirements and Constraints
    • Latency, Throughput, Capacity, and Bottlenecks
    • Scaling, Load Balancing, and Stateless Services
    Follow the system-design learning path Explore System Design
  7. AI for Developers

    Understand the software primitives behind useful AI systems and developer tools.

    • Tokens and context windows
    • Embeddings and RAG
    • Agents and MCP
    Explore AI for Developers

03 · Companion path

Software Development in the AI Era

A focused guide to using coding assistants while retaining the judgment to review, test, and ship their output. Separate from the software-fundamentals curriculum.

Eight concise lessons

Start with what an assistant can do. Continue through code review, testing, data boundaries, useful context, reliable workflows, and honest measurement.

Open the AI-era learning path

04 · Behind the Technology

Current conversations, explained through durable foundations.

A forthcoming editorial series about what developers are discussing, how the technology works, and what is useful beyond the hype.

First breakdown coming soon

Important developer conversations are easier to evaluate when you can see the durable systems beneath them. This series will separate facts, analysis, and opinion—then connect each discussion back to the foundations that explain it.

About the series

Example questions—not published articles

  1. What actually makes software an AI agent?

    AI systems · tools · control flow
  2. Why can a reverse proxy return a 502 error?

    HTTP · proxies · failure boundaries
  3. What happens behind an AI tool call?

    APIs · schemas · model orchestration

05 · Concept map

Understanding grows through connections.

A request depends on a chain of ideas. Learn the chain and debugging becomes a process of locating where the model broke.

  1. DNSFind the destination
  2. HTTPDefine the exchange
  3. APIsShape the interface
  4. AuthenticationEstablish identity
  5. CachingAvoid repeated work
  6. ScalingHandle growing demand