Book author. Google Developer Expert. Founder of angular.schule.
And working with AI agents every day for the past three years.
92 workshops of his own since 2016grown out of angular.schule
Which format suits your team?
Same workshop, same content. You choose the split.
The on-site team training at your company
Your team builds AI features you can stand behind –
on site at your company, in two to four days.
LLM APIs, agent architectures, MCP and AI interaction beyond the chat window. The base is two days on site or three half-days online, in which your team builds a feature from the API to the safeguards. On top you add the modules that fit your product: MCP server, RAG, agentic frontend, local models or n8n. We clarify the starting point and the focus in the intro call.
We come to you
The trainer travels to your location.
Travel, expenses and accommodation are included in the offer.
Full-day sessions
Two to four days of 7 hours each, e.g. 9am to 5pm incl. breaks. When we come to you, the workshop has the full attention of your team.
Your preferred date
We align the date together with your team.
A focused learning environment
Out of the daily grind, into full workshop mode:
no distractions, full focus.
You provide the setup
You provide a training room, a projector and a laptop per participant.
We bring the knowledge and the example code.
Code review beforehand
We look into your project beforehand: a code review shows where AI features can dock on and which modules are worthwhile.
2–4 days of AI training
On-site
From the API to a feature you can stand behind.
The base for everyone – plus the modules that fit your product.
As on-site training: 2 to 4 days of 7 hours each.
Topics & contents
BLOCK01- Models, providers, costs: the right choice for your feature
- Integrating LLM APIs: streaming, structured output, tool use Hands-on
- Prompts as code: versioning, tests, evaluation
- Error cases, timeouts and fallbacks
- Controlling latency and cost in production
- The first AI feature in the sample app
From the API to the feature
Integrating LLM APIs properly
An AI feature is only a feature once it answers reliably, fails cleanly and stays within budget. In the first block you lay the groundwork: from model choice via API integration to the first feature running in a real application.
- When an agent, when a workflow?
- Designing tools: interfaces a model understands
- State, memory and context limits
- Testing non-deterministic features: evals instead of gut feeling
- Guardrails: prompt injection, permissions, output control
- Observability: tracing, cost and quality in production
- An agent for a task from your domain Hands-on
BLOCK02Agent architectures
From single request to agent
Agents plan, call tools and decide on the next step themselves – that takes architecture. You learn to cut tools, manage state, and test and secure non-deterministic features so you can stand behind them.
MODULES03+- MCP server as a product interface
- Agentic frontend: AI interaction beyond the chat window
- RAG & your own knowledge
- Sovereign AI: local models on your own hardware
- Automating workflows with n8n
Your modules
You put the rest together
On top of the base you add what fits your product. Each module extends the agenda by four or eight hours; you choose in the configurator below. Order and examples are tailored to your project in the intro call.
Extend your training
Adapt the agenda to your product:
The Build for AI base consists of two on-site days of seven hours each. A deep dive extends the training by another on-site day, a compact module by half a day. We agree the exact schedule in the intro call.
All participants receive a certificate plus the sample code, setups and cheat sheets from the course to take home.
The author himself
Your trainer Johannes Hoppe is a Google Developer Expert (GDE)and co-author of the German-language standard reference
„Angular: The Practical Guide – from the basics to advanced development with Signals".
He runs every training himself – with the knowledge from his ongoing projects.
Johannes Hoppe
is a Google Developer Expert (GDE) for Angular and works as a freelance software architect, consultant and trainer focusing on agentic coding, Angular, Node.js and C# / .NET. Outside of teaching, he is a lead frontend developer at a mid-sized company.
Johannes lives & works in Leimen near Heidelberg.
Bluesky: @johanneshoppe.de
LinkedIn: @johanneshoppe
X (Twitter): @JohannesHoppe
Frequently asked questions – on-site team training
◢ What does the training cost?
The base costs €5,500 net on site (2 days of 7 hours) and €4,400 net online (3 half-days of 5 hours) – each as a fixed price for your whole team, on site including travel, expenses and accommodation. With twelve participants that is €459 or €367 per person. Each module is added at the same rate: a deep dive (8 hours) like another day, a compact module (4 hours) like half a day – €2,750 or €1,375 on site, €2,200 or €1,100 online. The full programme with all modules therefore costs €11,000 on site or €8,800 online. The price is per team, not per head. We agree the focus in a free intro call beforehand, and you receive the quote within 24 hours.
◢ What goes into the price?
We work with LLM APIs and agents every day, in our own products and in client projects, we tailor the focus to your product in the intro call, and at the end your team has built a feature it can stand behind. Measured against the months a team otherwise spends figuring out model choice, architecture and safeguards on its own, this is the cheaper option.
◢ Who is the training for?
Development teams that want to bring AI features into their own product – web, backend or both. You do not need prior AI knowledge, but programming experience: the examples are TypeScript-oriented, the concepts apply to any stack.
◢ What are the prerequisites?
Programming experience in the team. Prior knowledge of LLMs shortens the introduction; gaps we catch. What your team brings and where it wants to go, we clarify in the intro call.
On-site you provide a training room, a projector and a PC/laptop workstation per participant (with internet access). We send installation instructions for Node.js and the IDE ahead of time.
◢ Which models and SDKs are covered?
We work with the official SDKs of the major model providers and with open models via Ollama or vLLM – model interchangeability is an architecture principle for us. If only one specific provider is allowed at your company, we align the exercises accordingly.
◢ How hands-on is the training?
The course is a mix of theory and practice: together we develop a sample project into which you build the AI features step by step. Each topic is explained first and then implemented in code; individual tasks you work on in small groups. The code for the sample project is available on GitHub.
◢ Are there learning materials?
For every topic block we prepare a handout. Our detailed slide decks let you follow every topic comfortably. You get all the slides after the training. You also take home the sample code, setups and cheat sheets from the course – ready to use in your own project.
◢ Who pays for travel?
Travel, expenses and accommodation of the trainer are already included in the offer. You get a transparent flat fee. Whether the training happens in Munich, Hamburg or on the outskirts, there are no extra costs for you.
◢ What about catering?
You normally organise catering yourselves, your participants know their preferences best. We recommend cold drinks, coffee/tea, snacks for the morning break and a shared lunch, whether that is the cafeteria, a nearby restaurant or catering.
◢ What does the daily schedule look like?
On-site we typically run the workshop as full days, from 9am to 5pm, incl. a one-hour lunch break. If you prefer a different structure we'll agree on that together.
◢ Who is the trainer?
The training is run by book author and Google Developer Expert (GDE) Johannes Hoppe. He works in real projects and passes on his practical knowledge directly. Everything from a single source.
◢ Which language is the workshop held in?
In German or English, as you prefer; the course materials are in German, sample code and cheat sheets in English.
◢ What team size do you recommend?
We recommend a group of 8 to 12 participants, with a maximum of 18. From about 15 people upwards, the time available per question drops noticeably; in that case we suggest running two separate sessions.
◢ Why with us?
We tailor every team workshop to your product, your prior knowledge and your goals beforehand: send us code or architecture sketches and we build the agenda from them. Starting from a green field, we develop it during the course. We draw on the topics of more than 90 workshops and respond to whatever your team runs into.
◢ How can we book?
Put your modules together in the configurator and send the enquiry – you will receive a binding quote with all details within 24 hours. Or simply give us a call.
Still have questions?
+49 151 / 681 222 66 (Johannes Hoppe)
Standard rate to German mobile networksWe'll get back to you within 24 hours.Ready? Put your programme together.
The base plus your modules – you receive a binding fixed-price quote within 24 hours.
- Choose modulesIn the configurator you tick what fits your product and send the enquiry.
- Intro call and quote30 minutes, free of charge: we clarify the starting point and the focus. The quote arrives within 24 hours.
- A date that suits your calendarThere is no fixed course date. You say when – online or on site at your company.