TRAINING
AI Agents Expert
Through live sessions, real workflows, and hands-on designs, you’ll learn how to build, optimize, and deploy autonomous software systems that can perceive environments, plan actions, use tools, and execute multi-step workflows safely in the real world.
Start dates
10 November 2026
Duration
24 hours instructor-led and 24 hours self-paced training over 6 weeks
Course fee
CHF 1’990
Level
Expert
Language
English
Certificate
AI Agents Expert
Testimonials
Who is this training for?
Your next step starts here
AI agents are quickly becoming part of how real work gets done. This training is for professionals who already understand the basics of AI and want to go further, building automations and agents that connect to the tools they use, run safely, and produce useful results. You don’t need to be a developer. If you’re comfortable with core AI concepts and writing clear prompts, and you’re ready to start building, this is where you continue.
What you need to get started
This is an intermediate training that builds on what you learn earlier in your AI journey. Before you start, you should have previous education in AI literacy and AI prompt engineering, so you’re comfortable with core AI concepts and writing effective prompts. No coding background is required. A computer and a stable internet connection are all you need, along with a willingness to learn by building.
1. Visual workflows and your first AI calls
You’ll start hands-on with n8n, a visual tool for building automations you can see and control. You’ll learn how data moves between steps as structured information, how to wire a language model into a workflow, and how to use AI as a thinking partner that helps you sketch and refine what you build. By the end of this block, you’ll have built a working example and started spotting the tasks in your own day-to-day work that are ready to automate.
2. From prompts to real AI agents
You’ll move past one-off prompts and learn what turns a language model into an agent: a model working in a loop, with tools, memory, and your oversight. You’ll connect agents to the services people rely on every day, setting up the credentials that let them read email, write to a spreadsheet, or call a model directly. Along the way you’ll build the reflex that matters most: knowing how to authenticate each new integration, and why one method fits better than another. By the end, you’ll have a working setup ready to power agents that do real work.
3. Agentic coding and agentic work
This block is about working alongside agents instead of doing every step yourself. You’ll meet two agentic tools, Claude Code and OpenAI Codex, and make the mindset shift they ask for: you stop owning every step of the loop and start owning the review. You’ll use them to read, edit, and cleanly document your workflows, and even to handle non-coding work like filtering and ranking ideas against clear criteria. By the end, you’ll be able to hand real tasks to an agent and trust the result, because you’ll know exactly how to check it.
4. AI privacy, security, and governance
The most important questions about AI agents are not only technical. They are about trust. You’ll learn what can leak when an agent reaches out to an external service, and the everyday habits that keep your hand on the wheel: tighter defaults first, removing sensitive details before they cross a boundary, and verifying outputs rather than trusting a green checkmark. You’ll also see how to route each task to the right model to keep costs in check, and how frameworks like the EU AI Act and Swiss rules shape what builders are expected to do. By the end, you’ll be able to run agents that are safer, more compliant, and easier to stand behind in front of a customer or a regulator.
5. Front-ends for agentic workflows
A workflow only you can run is a tool no one else has. In this block you’ll learn to put a usable face on your work, turning a workflow into a proper app or service a colleague, client, or customer can run on their own. You’ll explore several ways to do it, from a simple built-in form to a hosted page to a fully custom interface, and learn to choose the simplest one that does the job. You’ll also see how to add a secure sign-in without building it yourself, using trusted providers instead. By the end, you’ll be able to ship something real, not just demonstrate it.
6. Capstone: build and ship your own agent
Everything you’ve learned comes together in a final build. You’ll take an idea from your own work and turn it into a working agent: a workflow with real integrations, sensible safeguards, and a face someone else can use. You’ll draw on the judgment you’ve built across the program, choosing the right tools, the right level of oversight, and the simplest path to a result. You’ll leave not just understanding AI agents, but with something you’ve built, run, and can keep improving.
What you’ll learn
Six blocks.
How you will learn
Our programs are built on learning science, not convention. Designed around how adults actually learn: through challenge, reflection, and direct application to real work. Delivered fully online, combining live sessions, hands-on activities, and structured independent work. Every session is recorded. Learning fits around you, not the other way around.
01. DISCOVER
Understand who you are and what you want. We start by mapping your goals, your context, and your experience, so everything that follows is built what around actually matters to you.
02. CO-CREATE
Shape the learning, not just receive it. Content and challenges are built around real professional contexts, so the learning feels relevant from day one.
03. APPLY
This is where knowing becomes doing. Real scenarios, live problem-solving, and hands-one exercises that connect directly to your professional experience.
04. REFLECT
Connect insights, strengthen what lasts. Through discussion and guided debriefs, you link new knowledge to what you already know and make it yours.
05. IMPACT
You leave with more than knowledge. You leave with the confidence, judgement, and capability to make a difference where it counts.
Learning activities
Every session is supported by practical resources that help you prepare, review, and apply what you learn between live classes.
Flashcards
Bite-sized concept reviews to reinforce key terms and ideas between sessions.
Quizzes
Short knowledge checks that help you track your understanding as you progress.
Cheatsheets
Concise reference guides you can use during sessions and take back to your work.
Session plans
A clear overview of each session so you know what to expect and how to prepare.
Reflection cards
Structured prompts that help you connect what you have learned to your own professional context.
Further readings
Curated resources for those who want to go deeper on any topic covered in the program.
Assessment
Designed around real professional practice
Assessment at SCI is designed to reflect real competence, not the ability to perform under artificial pressure. Before your program begins, you will know exactly what you are being assessed on, how, and against what standard. Progress is tracked continuously through practical challenges and applied scenarios built into the learning itself.
What you demonstrate at the end is not recall. It is capability.
Investment overview
Training fee
CHF 1’990
Certificate
AI Agents Expert
You will be provided the “AI Agents Expert” certificate in an electronic format.
Corporate Training
Upskilling in AI
Is your team interested in the AI Agents training? Let’s talk.
Your instructors
Petr Lavrov
Instructor
Petr is an AI, productivity, and automation enthusiast with a strong background in software engineering, having worked at leading tech companies such as Superlinear, Google, Magic Leap, and Yandex. A passionate problem solver, he is an international award winner in mathematics, physics, and programming.
Nabil Alouani
Instructor
Nabil is a project manager and an expert in observability and data tooling for distributed systems serving 80,000+ daily users. He brings real-world depth in real-time monitoring pipelines, AI-powered anomaly detection, LLM-based analytics, agentic workflows, and RAG.
Book a free consultation call
Our students work at
Our courses
Subscribe for updates
We share the most important news from the industry, technology, and the institute. Stay informed and stay ahead.
