ai-agents
13 free lessons tagged ai-agents across AI, Business. Each one is a short sequence of focused steps with narration and a five-question quiz at the end — take them in any order, no signup required.
Measuring a Harness Change: Trials, Error Bars and pass^k
Every harness change is a hypothesis. Learn how evaluation harnesses record agent runs, why one run is one noisy sample, how paired comparisons and repeated trials separate real gains from luck, and when to report pass@k versus pass^k, with computed simulations.
Permissions, Sandboxes and Hooks: Containing an AI Agent
An agent with a shell can make destructive mistakes or follow injected instructions. Learn the four layers of control, from instruction files to OS sandboxes, why command-pattern rules are not a boundary, how hooks add deterministic checks, and how to design out the lethal trifecta.
The Repository as the Agent's Environment
Coding agents only know what the repository shows them. Learn to write instruction files as maps, turn conventions into checks with fix-it messages, give agents sensors, and structure work that spans many sessions, using OpenAI's, Anthropic's and ETH Zurich's findings.
Model Plus Harness: Why the Same Model Scores Differently
An agent is a model plus everything around it. Using controlled evidence from SWE-agent and Terminal-Bench, see how interface design, harness choice and even container limits move the results of a fixed model, and why the model still sets the ceiling.
Agentic Commerce: Trust, Liability, and What Breaks
Scoped tokens and signed mandates solve authorization, not everything. This lesson covers the unsettled parts: who is liable when an agent buys wrong, why merchants and agents have opposing incentives, prompt injection as a new attack surface at checkout, agent-to-agent buying, and why the market forecasts disagree by an order of magnitude.
Delegating Spending Authority to an Agent
How do you let software spend your money without handing it your card? This lesson covers the mechanism behind agentic checkout: scoped single-use credentials like the Shared Payment Token, Google's AP2 intent and cart mandates as signed verifiable credentials, agent identification, and the guardrails that bound what an agent may buy.
Building MCP Servers in Python and TypeScript
Learn to build Model Context Protocol servers from scratch using the official Python and TypeScript SDKs. Cover tool schemas, resource URIs, prompt templates, and how Claude Code, Claude.ai, and Cursor consume them.
Model Context Protocol: The Open Standard for AI Tool Integration
A deep dive into MCP — Anthropic's open protocol for connecting LLMs to external tools, data, and prompts. Covers JSON-RPC transport layers, the three core primitives, capability negotiation, and how to build a working server from scratch.
Agentic RPA and Computer-Use Agents: Benchmarks vs. Real Deployments
What measurable benchmarks say agentic software agents can actually do in 2026, what production deployments have publicly disclosed, the strongest skeptic case on the books, and how the EU AI Act reshapes the deployment math.
AI Agents in Scientific R&D: AlphaFold, Insilico, and the Reproducibility Bar
Where the evidence for AI in scientific discovery is strongest, where it is weakest, and how to tell them apart. Anchored on AlphaFold's peer-reviewed record, the 2024 Nobel Prize in Chemistry, and what AI-first drug-discovery companies have actually disclosed to investors.
Legal AI Agents: Adoption Patterns and Documented Failure Modes
Evidence on AI in legal practice: the Stanford RegLab hallucination benchmark of Westlaw and Lexis+ AI, the Mata v. Avianca and Park v. Kim sanctions cases, and what published ethics rules require.
Coding Agents: Productivity Studies, Benchmarks, and the METR Slowdown
Hard evidence on AI coding assistants: the Microsoft Research RCTs, METR's surprising 2025 slowdown finding, SWE-bench Verified, and why benchmark scores keep outrunning real-world value.
Customer Service AI Agents: What the Data Actually Shows
An evidence-led look at AI agents in customer support: the Brynjolfsson NBER RCT, Klarna's 2024 victory lap and 2025 walk-back, and what generalises beyond a single firm.