Autonomous AI agents in your company
Build an agent prototype that works with your company's files, knowledge and tools
- Exercises in every course
- Final exam
- AI mentor
Learn to delegate tasks to AI agents in your company.
In four workshops you will build an agent prototype that works with files, knowledge and tools. You will learn to check its results, access scope and running cost.
You work with Claude Code and Codex in a prepared lab. You don't need to program yourself.
You finish with a working prototype, its operating instructions and a plan for using it in your own company.
- 4 live online workshops, 3 hours each
- Exercises in the lab on edu.aiversum.ai
- A reporting agent prototype on training data
- Assessment of the final project
- Recordings and materials for 90 days
- Questions to the instructor for 30 days after the course
Practical skills backed by a certificate
All training runs on our EDU platform. You practise real tasks, reinforce your knowledge, take an exam and receive a certificate. Your company can track the team’s progress.
- Exercises in every course
Every course has exercises on real tasks, quizzes and decisions during lessons. You learn by doing.
- An exam after every course
At the end you check what you can really do. The result goes into your progress.
- Reels: what you can use today
Short vertical videos, each with one thing to use at work today.
- Reminders and reviews
The platform reminds you about exercises and reviews so the new skill becomes a habit.
- AI mentor
An AI assistant answers questions about the material and helps with exercises when you get stuck.
- Progress your company can see
Your company sees who has made what progress: lessons, exercises and exam results.
What you'll learn
Task and result
- How to describe a task so an agent can do it
- Correctness criteria: how you recognise a good result
Context and repeatability
- A knowledge and instruction pack for the agent
- A skill: a task the agent repeats on new data
Tools and autonomy
- CLI and MCP tools: how an agent reaches systems
- Access scope and handling missing information
Control and application
- Execution trace, cost and result tests
- A plan for using the agent in your company
Programme
- Session 1 · Task and result
You describe the task, prepare the data and run the first analysis. You define correctness criteria.
- Session 2 · Context and repeatability
You build a small knowledge and instruction pack. You prepare a skill and run the task on a new dataset.
- Session 3 · Tools and autonomy
You configure a prepared CLI or MCP tool, the access scope and handling of missing information.
- Session 4 · Control and application
You test the report, review the execution trace and cost. You see agents working together and write a plan for your company.
- After the course
Project assessment, recordings for 90 days and questions to the instructor for 30 days.
Instructors
Builds aiv, a working environment for AI agents in companies, and deploys agents in client processes. Works with Claude Code and Codex every day. Runs AI and security training and webinars for IT teams.
Who this is for
You are responsible for tools and want to know what an agent can safely do in your systems.
You work with data and reports. You want an agent to prepare them repeatably and verifiably.
You run AI initiatives and need practice, not another presentation.
What's included
- Live sessions
Four online sessions with the instructor, in a small group.
- Lab
A prepared environment with a model usage limit. You don't configure anything yourself.
- Recordings and materials
Access for 90 days after the last session.
- Feedback
Assessment of the final project and answers to questions for 30 days.
Frequently asked questions
Do I need to know how to program?
No. You work in a prepared environment. Being comfortable with files, data and applications is enough.
What if I miss a session?
Every session is recorded. You have the recordings and materials for 90 days.
How much time does it take?
Four 3-hour sessions plus exercises between sessions.
Can my company pay for the course?
Yes. We issue a VAT invoice and prepare documents for the procurement process. You can book several seats for your team.
Will we connect the agent to my company's systems?
The course ends with a prototype on training data. Connecting to production systems is a separate project we can discuss after the course.
For companies
VAT invoice and procurement documents. We'll help you justify the spend.
Request documentsThe course teaches you to work with agents. When you want an agent running on a company process, on your systems and within your security boundaries, we build it with you.
See how we work