Practical guides to building AI agents in Python: tools and MCP, memory, guardrails, approvals, observability, testing and deployment, each with runnable code. For everyone else: using AI at work, leading AI adoption and enterprise AI agents. Free, no account needed.
01 · Guides
Standalone articles. Each one explains a single technique or idea, with worked examples you can try.
02 · Learning paths
Short courses made of guides in a set order, with what to know first and what you can do after.
03 · Levels
Every guide is marked Beginner, Intermediate or Advanced, so you can start where you are.
Start where you are
I use AI at work →
Prompts that get usable results, safe use of company data, checking answers, role playbooks.
I lead a team →
What AI agents can do, choosing use cases, build vs buy, rollout and ROI.
I evaluate AI for my company →
Maturity model, reference architecture, platform evaluation and governance.
I build AI agents →
Tools and MCP, memory, guardrails, approvals, runtime and deployment, with code.
Secure prompt engineering: keep untrusted text as data with message roles, delimiters, spotlighting, rules and output schemas. A real before and after.
LLM supply chain security as a checklist: inventory models, data and tools, pin and verify hashes, load safe formats, check provenance and plan fallbacks.
The LLM attack surface explained: user apps, integrations, APIs, data stores, insiders and supply chain, the common attack vectors, and the OWASP Top 10.
LLM agent orchestration patterns: single agent, chaining, routing, supervisor-workers and handoffs. When each fits and what the controller must enforce.
Build a ReAct agent from scratch in Python: the Thought, Action, Observation loop, a 60-line controller with a real trace, and how it maps to tool calling.
What prompt engineering is, how a prompt works, and the five parts of a good one, with a real before-and-after example and a map of the beginner path.
Each topic page explains the idea, then lists its guides in the order to read them, with links to the matching documentation.
Categories
Topics
A learning path is a set of guides in a fixed order, like a short course. Each one says what to know before you start, the skills you build and what you can do at the end.
This path teaches you to design prompts that do more than generate answers. They reason, verify, adapt and collaborate. You’ll learn how to build workflows that reduce errors, create consistency, and scale across complex tasks. Each guide combines theory with practice, so you’ll not only understand the techniques but also test them in small labs that make the difference visible. By the end, you’ll know how to turn a single model into a reliable system: one that can explain its steps, double-check its own work, and even coordinate multiple reasoning strategies before deciding on the best answer.
The complete path to building AI agents in Python with Promptise Foundry, from your first agent to production. You start with what an agent is and how to build one, give it tools through the Model Context Protocol, add memory and retrieval, and choose reasoning patterns. Then you make it safe and dependable: guardrails, prompt injection defense, human approvals, tracing and debugging, cost limits and evals. The path ends with running agents in production: triggers, crash recovery, deployment and the promptise run project runner. Every guide explains one concept and shows runnable code.
A free, practical course for anyone who uses ChatGPT, Claude, Gemini or Copilot at work. You learn to write prompts that get usable results for emails, summaries, meeting notes, reports and analysis; to keep company data safe; and to check AI answers before you rely on them. Role playbooks show the same skills applied to HR, marketing, sales and finance. No technical background needed.
For team leads, managers and executives who decide where AI goes in their organization. You learn what AI agents can and cannot do today, how to pick use cases with a scorecard, when to build and when to buy, how to roll AI out to a team in 90 days, and how to measure whether it pays off. Every guide ends with a tool you can copy: a scorecard, a decision table, a rollout plan or an ROI worksheet.
For IT, security, platform and architecture leads bringing AI agents into a large organization. You assess where you stand with a maturity model, learn a vendor-neutral reference architecture, evaluate platforms with a weighted scoring sheet, and set up operational governance: owners, approvals, limits, logging and reviews. The path ends with what an agent runtime is and what to check before an agent goes live.
This path takes a true beginner from “I asked… it guessed” to “I ask clearly and get dependable results.” You’ll learn why language models follow structure more than intent, how to give them the right context, and how to lock the shape of outputs so they’re paste-ready. Along the way you’ll practice quick checks that reduce hallucinations, add gentle reasoning, and shape tone without losing accuracy. The last modules introduce multimodal basics and a tiny step into retrieval so you can ground answers in your own material. Everything is hands-on, narrative-first, and easy to reuse at work.
A free LLM security course for developers and security teams, from how LLM attacks work to defending production systems. You learn the main attack vectors (prompt injection, data leakage, unsafe output), how to structure prompts and validate input and output, how to red-team your own system, and how to monitor and respond when something goes wrong. Each lesson is a hands-on guide with examples you can run.
This path helps you shift from using prompts to thinking like a prompt engineer. Instead of teaching techniques, it introduces mental models. These are ways of seeing and reasoning about large language models. Each guide is a mindset essay with visual diagrams and reflection prompts that sharpen your intuition, inspire creativity, and build the resilience needed to invent new solutions. By the end, you’ll think like a builder, a critic, and a collaborator with AI.
The guides teach the techniques. Promptise Foundry, our open-source Python framework, turns them into software: typed prompts, governed agents, real MCP servers.