Rinoova
Back to blog
AcademyFormazioneAI

Free AI courses with a certificate: 8 new courses

Eight new free courses on Claude Code, AI agents, prompt engineering, evals, security and AI for small businesses. No sign-up, verifiable certificate.

Share

In July we launched Rinoova Academy with a single free course, Model Context Protocol: Fundamentals. Today there are nine free courses: we've added eight new ones, ranging from automation in small businesses to AI agent security.

The rules are the same for all of them:

  • Entirely free: content, final test and certificate.
  • No sign-up: open the course and start reading.
  • About two hours each, at your own pace.
  • Verifiable certificate: pass the final test with at least 80% correct answers and you get a Certificate of Attendance with a code and a public verification page. It's an Open Badge 2.0 you can add to your LinkedIn profile.
  • In English, Italian and French.

See all free courses

Here are the new courses, grouped by who they're for.

For people running a business

AI Automation for Small Businesses: how to bring AI automation into a company without wasting budget. It starts from mapping repetitive work and goes on to picking the first process, deciding between classic rules, AI or both, calculating the real return, designing human oversight and moving from pilot to production. The last chapter is about people: training and adoption.

AI and Company Data: Protecting Your Information: what happens to your data when your company uses AI tools. Where the data you send ends up, how to classify it before using it, how to choose a vendor (retention, training, region), when to involve whoever handles privacy, and how to manage access, logs and ownership of generated content.

For developers

Production-Ready Prompt Engineering: prompts that work in production, not just in demos. Structure, context, few-shot examples, validated JSON output, edge cases, prompts for systems with tools and document retrieval, and how to version and maintain them over time.

Claude Code: a practical guide: using a command-line coding agent without losing control. CLAUDE.md, permissions, plan mode, commands, hooks and MCP servers, subagents and worktrees for parallel work, and a daily workflow built on tests, small commits and review.

AI agents: how they actually work: from a model that answers to a system that acts. The agentic loop (observe, decide, act, verify), how to design tools, memory and context, long-running tasks, and when using several agents together makes sense and when it doesn't.

Context engineering: what goes into a model's context: deciding what a model or agent sees, when, and what stays out. What context costs, how to retrieve the right documents, how to summarise without losing information, how to split work into sub-agents and what to keep in memory between sessions.

For people putting AI into production

Evaluating AI systems (evals): how to measure whether a language-model-based system really works, before and after release. Criteria and error severity, sets of real, edge and adversarial cases, automated checks, a model as judge calibrated against human judgement, and regression testing on every prompt or model change.

Securing AI agents and MCP servers: what to check before connecting an agent to company systems. Direct and indirect prompt injection, untrusted tools and MCP servers, least privilege, tokens and the confused deputy problem on remote servers, isolation, human confirmation, logging and incident response. It follows the 2026-07-28 revision of the MCP specification and the 2025 OWASP Top 10 for LLM applications.

Where to start

If you're not sure which course to take first, here are three possible paths:

  1. Business owners and department heads: AI Automation for Small Businesses, then AI and Company Data.
  2. Developers: Production-Ready Prompt Engineering, then Claude Code, AI agents and Context engineering.
  3. Architects, platform and security teams: Model Context Protocol: Fundamentals, then Securing AI agents and MCP servers and Evaluating AI systems.

The courses are independent: you can take just one and get its certificate.

If MCP is what you're after, the blog also has a hands-on example: how to turn existing APIs into an MCP server.

After the free courses

If you want to go deeper, there are the limited-seat executive masters: MCP Architect, AI Agent Engineering and AI Coding Orchestration. For teams there's in-company training, done on your own codebase: we covered it in the article introducing the Academy.

For a programme tailored to your company, get in touch.

Frequently asked questions

Are Rinoova Academy courses really free?

Yes. The content, the final test and the certificate are entirely free, with no paid modules and no request for a payment method.

Do I need to sign up to take the courses?

No. You can read the courses without signing up. Your name and email are only needed to receive the certificate, after you pass the final test.

How do I get the certificate?

By passing the final test with at least 80% correct answers. The certificate is issued automatically, has a code and a public verification page, and is an Open Badge 2.0 you can add to LinkedIn. If you don't pass, you can retake the test.

How long is each course?

About two hours each. They are text-based courses you read at your own pace.

Which languages are the courses available in?

English, Italian and French.

Which course should I start with?

It depends on your role. Business owners can start with AI Automation for Small Businesses, developers with Production-Ready Prompt Engineering or Claude Code, and whoever has to put a system into production with Evaluating AI systems and Securing AI agents and MCP servers.

Share