Primers

Get oriented on the AI stack, fast. Each primer covers one tool or technology in under 25 minutes — what it is, why it matters right now, and how to get started. Every primer is dated and re-verified as the ecosystem moves.

Coding Agents 4 Agent Building Model Context Protocol 2 Local & Open Models 1 Voice & Multimodal Evals & Observability 1 RAG & Retrieval 1 AI-Native IDEs Browser & Computer Use Deployment
Coding Agents Intermediate

Opus 4.7: Task Budgets and the Memory Tool

What actually changed in Claude Opus 4.7 — the task budget system, the persistent memory tool, and how they combine to make long-horizon agents finally reliable.

14 min read · Verified Apr 2026
Coding Agents Intermediate

Caveman Claude: the skill everyone's talking about

What the viral Caveman skill actually does, why a one-paragraph prompt trick ate developer Twitter this month, and whether the 75%-fewer-tokens claim survives contact with real work.

15 min read · Verified Apr 2026
Model Context Protocol Foundational

MCP in 20 Minutes

What the Model Context Protocol is, why it matters now that it's vendor-neutral, and how to connect your first MCP server from Claude Code, Cursor, or ChatGPT in under ten minutes.

18 min read · Verified Apr 2026
Evals & Observability Intermediate

Your first eval suite with Braintrust

The 60-minute path from 'we ship LLM features on vibes' to 'we block merges on eval regressions'. Datasets, scorers, experiments, and the CI wiring that makes it real.

15 min read · Verified Apr 2026
Coding Agents Foundational

Prompt Caching: 90% Cost Reduction in 5 Minutes

How Anthropic's prompt caching works, why the February 2026 workspace-isolation update made it finally safe for multi-tenant products, and the exact code changes to cut your API bill by an order of magnitude.

12 min read · Verified Apr 2026
Local & Open Models Intermediate

Local coding on an M4 Mac: Ollama + Devstral

When a local model earns its slot. The full setup on Apple Silicon — Ollama, Devstral Small 24B, Metal acceleration, and the editor integrations that make local coding stop feeling like a demo.

12 min read · Verified Apr 2026
RAG & Retrieval Intermediate

pgvector: why most RAG should just be Postgres

The contrarian case. pgvector + pgvectorscale gets you HNSW-indexed vector search inside the database you already operate. When a dedicated vector DB is worth it, and when it's just operational debt.

14 min read · Verified Apr 2026
Model Context Protocol Foundational

Build your first MCP server in TypeScript

Ten minutes from zero to an MCP server exposing a useful tool to Claude Code, Cursor, or any MCP client. The SDK, the transports, the gotchas.

14 min read · Verified Apr 2026
Coding Agents Foundational

Claude Code from zero to daily driver

The path from first install to 'I use this for everything': plan mode, hooks, skills, MCP, subagents, settings hierarchy, and the habits that actually stick.

18 min read · Verified Apr 2026