Lukas Kaminskis
Course creator
Co-founder & CEO of Turing College, backed by Y Combinator
88%*
of graduates get hired 6+ months after graduation
*Germany, as of July 2026. Includes learners whose programme ended at least six months earlier and whose employment status could be confirmed. Unconfirmed outcomes were excluded.
Become an AI power user
Aim for 5 to 10 times more output with AI.
Learn to set up agents that run workflows, test ideas and flag decisions for you. Choose your role to see what you could build.
Task
No AI
Marketer
You brief specialists, wait for their availability and coordinate the handoffs.
Advanced AI skills
AI Marketer
Build and run the workflow yourself with AI agents. You set the goal and approve.
How the work gets done
- You
- Analyst
- Creative
- Media buyer
- CRO
- Live
- You + AI agents
- Live
Campaign dashboard
A weekly export stitched together in a spreadsheet
Agents monitor spend, leads and revenue, flag shifts and suggest your next move.
Ad creative variants
A brief, a review round and a queue for new variants
Agents create variants and check them against your brand rules. You approve the shortlist.
Landing page test
Mockup, ticket, build, QA, then wait for a release window
Set up agents to run A/B tests and analyse results. You choose what to roll out.
No AI
Marketer
You brief specialists, wait for their availability and coordinate the handoffs.
How the work gets done
- You
- Analyst
- Creative
- Media buyer
- CRO
- Live
Campaign dashboard
A weekly export stitched together in a spreadsheet
Ad creative variants
A brief, a review round and a queue for new variants
Landing page test
Mockup, ticket, build, QA, then wait for a release window
Advanced AI skills
AI Marketer
Build and run the workflow yourself with AI agents. You set the goal and approve.
How the work gets done
- You + AI agents
- Live
Campaign dashboard
Agents monitor spend, leads and revenue, flag shifts and suggest your next move.
Ad creative variants
Agents create variants and check them against your brand rules. You approve the shortlist.
Landing page test
Set up agents to run A/B tests and analyse results. You choose what to roll out.
5 to 10 times more output is an ambition for repeatable tasks, not a measured course result. Your gains depend on the workflow and how you use AI.
Not sure which course to take?
AI Benchmark Quiz
How strong are your AI skills?
Take the quiz to see which Building with AI skills you already have, where your gaps are, and how the programme gets you to Black belt.
6–7 questions + 2 skills check-ins · about 3 minutes · email required
Example result · not your score
AI Marketer
Yellow belt
4/15
programme skills checked
Building with AI · Programme skill ladder
- Black beltProgramme completion level15
- Purple belt10–14
- Green belt7–9
- ExampleYellow belt3–6
- White belt0–2
4/15 skills checked
+3 skills → Green belt
What can you expect?
Learn, build, get feedback. Repeat with a more ambitious project.
Projects that go live
Follow a clear path from a simple web tool to an AI assistant and a final project based on your own idea.
One-to-one expert feedback
Review your work with experienced professionals. Understand what AI produced and how to improve it.
Tools introduced step by step
Learn Claude Code and supporting tools through projects, exactly when you need them. No technical background needed.
What you’ll be able to do
- Turn a business problem into a working tool
- Learn to give AI coding agents clear goals, useful context and repeatable instructions. Use specifications, agent memory and custom skills to guide their work.
- Build and launch dashboards, internal tools and websites
- Create complete tools using Next.js, Supabase and Vercel. Add databases, user accounts and connections to external services.
- Guide AI agents towards more reliable results
- Coordinate AI agents, connect them to tools through MCP and use automated browser tests to check whether the finished product behaves as expected.
- Take a prototype to a live product
- Add custom domains, logins, payments, testing and automated deployment so you can take what you build beyond an initial prototype.
Building with AI case study
From construction expertise to an AI product.
After 20 years in construction, Rimvydas Samulionis is building Ceiling Advisor through Building with AI: an assistant that helps architects and customers find answers in technical product documents, with sources they can check.
My role is to understand what needs to be built, make good decisions and make sure all the parts work together.
Rimvydas Samulionis
Project Manager at MONTEM · Building with AI learner
Optional: Watch the Doku build
Optional: Watch the Doku build
For another example, watch Lukas build Doku, Turing College's internal knowledge tool. Lukas had prior technical experience; this walkthrough shows his process.
What you'll build, sprint by sprint
Four sprints, each ending with a project that’s live online. Self-paced: most learners finish in 3.5 to 4.5 months at around 12 hours a week.
01Your first web tool, live online
01Your first web tool, live online
Set up Claude Code, learn to brief an AI agent so it builds what you mean, and save your work with Git and GitHub. You’ll put a small web tool online, then build your first Next.js app. Sprint project: your first Next.js app, live at a link you can share.
Briefing an AI agent clearlySaving your work with Git and GitHubA web app live online
Tools you’ll learn: Claude Code, Git, GitHub, Next.js.
02A tool your team can log into
02A tool your team can log into
Teach Claude Code your repeat tasks with slash commands, skills and plugins. Then add a Supabase database and user logins, and learn to fix errors with the agent’s help. Mid-sprint you’ll build a notes app with collections and search. Sprint project: a working app with its own database and user logins.
Reusable Claude Code skills and pluginsA live Supabase databaseUser loginsFixing errors with Claude Code
Tools you’ll learn: Claude Code, Supabase, Next.js.
03An AI tool that knows your data
03An AI tool that knows your data
Split bigger jobs between subagents, put your app online with Vercel and learn the security basics every app needs. Connect AI to other services with MCP, test your app automatically with Playwright, then build an AI tool that answers questions from your own documents. Sprint project: an AI app of your own, live online.
Your app online with VercelSecurity basicsAutomated tests with PlaywrightAI that answers from your own data
Tools you’ll learn: Claude Code, Vercel, MCP, Playwright, OpenRouter.
04A mobile app, payments and automations
04A mobile app, payments and automations
Automate recurring work with Claude Code, learn the compliance rules for AI tools and build a mobile app with an AI feature. Then plan your design before you build, add content management and payments, and launch your own online shop. Sprint project: your own online shop.
Automations with Claude CodeCompliance basicsA mobile app with an AI featurePayments and content management
Tools you’ll learn: Claude Code, Expo, Stripe.
CapstoneYour own idea, built and launched
CapstoneYour own idea, built and launched
Plan, build, test and deploy an AI-powered application of your own design, ideally the tool your team has been waiting for. Present it in a live demo and leave with a product you can keep developing after the course.
Your own working productA live demoA portfolio piece for your CV
Tools you’ll learn: Claude Code, Supabase, OpenRouter, Stripe, Vercel.
The tools you’ll learn
You don’t need to know these tools before you start. You’ll learn what each one does and when to use it through guided projects. You define what you need, guide Claude Code to build it and check that it works. Step by step, you’ll learn to build a tool, connect it to data and share it with a link.
Meet the people who built the programme
The programme is built and kept current by people who direct AI coding agents every day, develop AI products and lead software teams. They review it as the tools change, so what you learn is how it is actually done now.
Professionals like you, building for their own jobs
The content is clear, relevant, and links theory to real workplace scenarios. I've been able to apply what I'm learning straight away to work more efficiently, plan tasks better, and communicate more clearly with stakeholders.
I've built a stronger understanding of how AI fits into our processes. I've used what I've learned to speed up data tasks, improve reports, and spot opportunities where AI can save time for myself and other staff members.
I've learnt to use prompts in a more efficient and effective way. Recently, I used AI to identify gaps in our processes and produce clear meeting outcomes for the wider team. It was much faster than typing up notes and gave us a clear way forward.
The tutor meetings and workshops provide a great opportunity to discuss the learning. I've already used the prompt engineering and productivity tips to run my own training sessions for my colleagues, helping them use AI more effectively in their own work.
What Building with AI Agents learners build
During the programme, you will start with a problem you want to solve and finish with a tool that solves it. AI agents do the building, and you decide what they build and check the result.
Think about a task in your work that takes longer than it should, like a report you rebuild every week or information you copy between spreadsheets. That's the kind of problem you could solve with your own tool.
Here are some of the projects Turing College learners have built on the programme. Open any of them to see what it does and how it was made.
Find the right Turing College AI course for you
Compare the programmes to find one that fits your experience and the work you want to do.
The durations and weekly hours shown are for self-funded study.
Swipe to compare
This programme Building with AI Agents | AI Engineering | AI for Business | |
|---|---|---|---|
| Who it's for | Professionals in consulting, marketing, product, sales, finance and operations who want to build their own tools to solve business problems | Developers who want to use their coding skills to build, evaluate and deploy LLM applications and AI agents | Professionals in marketing, finance and operations who want to use AI to analyse information, create content and automate recurring tasks |
| Coding experience needed | None | At least 1 year of Python or JavaScript | None |
| Duration | 3.5 to 4.5 months | 3 to 4 months | 2 to 4 months |
| Hours per week | ~12 | ~10 | ~9 |
| What you'll build | Deployed tools for tasks such as managing documents or finding information, an AI assistant, and a final AI application you design and build yourself | Applications that answer questions using external data (RAG), agents that carry out multi-step tasks, and an independently designed AI application | Presentations, AI-assisted spreadsheet analyses and automated workflows that connect apps to summarise, organise or route information |
| Key tools |
Living in Germany? Building with AI Agents can be 100% funded with a Bildungsgutschein. Building with AI Agents with a Bildungsgutschein
How admissions works
Two simple steps, designed to help you and our team understand whether the programme is the right next step for you, and what you would build first.
- 1
Application
Tell us about your goals and the tool you want to build. Our admissions team reviews your application personally. - 2
Consultation
Meet us for 20 to 30 minutes to discuss your project, schedule and funding. Bring your benchmark result as a starting point.
Basic admission requirements:
No coding background
No coding background
None needed. You direct an AI coding agent in plain English; the programme explains technical ideas only when a project needs them.
Time commitment
Time commitment
Plan for around 12 hours per week. The programme is self-paced, and most learners finish in 3.5 to 4.5 months.
English proficiency
English proficiency
Materials, mentor sessions and project reviews are in English, and English is how you brief your AI agent. Reading comfort matters most; many learners speak English as a second language.
A computer and stable internet
A computer and stable internet
Any modern laptop works. Nothing to install that we don't walk you through, and no special hardware.
Flexible payment and funding options
You’re resident of:
Education voucher (Bildungsgutschein)
Best for
German residents eligible for a Bildungsgutschein
Course duration
Fixed, 3 months
Time commitment
30 h per week
Price
€0Fully funded
With a Bildungsgutschein
Claude Team Premium seat includedUpfront payment
Best for
Learners who want the lowest total price by paying upfront
Course duration
Flexible, 3.5 to 4.5 months
Time commitment
12 h per week
Price
€2,000Inaugural scholarship
Regular price €3,500
14-day money-back guaranteeInstalments
Best for
Learners who prefer to split self-payment into monthly instalments
Course duration
Flexible, 3.5 to 4.5 months
Time commitment
12 h per week
Price
€625× 4 monthsInaugural scholarship
Regular price €990 × 4 months
14-day money-back guaranteeUpfront payment
Best for
Learners who want the lowest total price by paying upfront
Course duration
Flexible, 3.5 to 4.5 months
Time commitment
12 h per week
Price
€2,000Inaugural scholarship
Regular price €3,500
14-day money-back guaranteeInstalments
Best for
Learners who prefer to split self-payment into monthly instalments
Course duration
Flexible, 3.5 to 4.5 months
Time commitment
12 h per week
Price
€625× 4 monthsInaugural scholarship
Regular price €990 × 4 months
14-day money-back guaranteeChoose your next group. Apply now!
Starting dates
November 9th
Limited seats
FAQ
What is Building with AI Agents?
What is Building with AI Agents?
Building with AI Agents is a beginner-level course for professionals who want to solve a real problem at work using AI. You direct AI to write the code in plain English, while you handle the planning, the context, and the final review. What you build can be as small as a dashboard that keeps itself up to date, or as ambitious as a full working product, like the internal tool our CEO built to replace a paid software subscription entirely by directing AI agents. Across four sprints, you'll build things like an internal tool, a dashboard, and an AI assistant, ending with a capstone project of your own choosing. No prior coding experience is required.
What is agentic coding, and what does an AI coding agent do?
What is agentic coding, and what does an AI coding agent do?
Agentic coding is a way of building software by giving an AI agent a goal and allowing it to complete a sequence of development tasks. Depending on the access you provide, an AI coding agent can inspect project files, write and edit code, run commands, test an application and respond to errors. You remain responsible for the goal, context, permissions and final review.
Why agentic coding is becoming a practical way to build
Why agentic coding is becoming a practical way to build
You decide what to build. The agent helps execute it.
AI coding agents can inspect a project, change files, run commands and tests, and work through a sequence of development tasks. The person directing the agent still makes the product decisions, supplies context and checks whether the result works as intended.
In June 2026, Anthropic published research based on roughly 400,000 Claude Code sessions ↗ recorded between October 2025 and April 2026. In a typical session, people made most of the planning decisions while Claude made most of the execution decisions. The research also found that people across major occupations completed coding tasks at close to the average success rate of software engineers, while relevant domain expertise improved the likelihood of success.
The 2025 DORA State of AI-assisted Software Development report ↗ reached a related conclusion for software teams: AI tends to amplify the strengths and weaknesses of the system around it. Better results depend on clear workflows, sound development practices and appropriate technical safeguards, not access to an AI tool alone.
This course teaches the workflow around the agent. You will learn how to plan, provide context, test, debug, review and deploy, so a quick prototype can become software you can explain and improve.
Do I need coding experience, and is this a no-code course?
Do I need coding experience, and is this a no-code course?
No prior coding experience is required. The course is built around Claude Code, the main AI coding agent you'll use throughout. If you already use or prefer a different agent, such as Codex, you can complete most of the course with it too.
You'll work with real, working code and real developer tools, not a closed drag-and-drop platform. The AI handles most of the actual coding, while you direct the work and check the results.
The course introduces technical concepts only when you need them. You should be comfortable learning new tools, checking results, and troubleshooting when something doesn't work as expected.
What will I build during the course?
What will I build during the course?
You'll build a working internal tool, an application with a live database and logins, an AI personal assistant, a mobile app, a subscription product with payments, and a capstone project of your own choosing. Past examples include Rimvydas, a project manager with 20 years in construction and no coding background, who built a tool that checks his company's product documentation for him instead of doing it by hand, and our CEO Lukas, who had prior technical experience and rebuilt an internal Notion alternative in around 30 hours.
How is Building with AI Agents different from vibe coding?
How is Building with AI Agents different from vibe coding?
Vibe coding usually means prompting an AI tool until a prototype appears to work. Building with AI Agents teaches a structured process for defining requirements, managing context, reviewing changes, testing behaviour, protecting data and deploying software. You still direct the agent, but you also learn how to check and improve what it produces.
Can I build a production application without being a developer?
Can I build a production application without being a developer?
AI coding agents can complete many implementation tasks, but a working prototype is only part of a production application. You also need to define requirements, manage access, test behaviour, protect data, check security and maintain the product. The course teaches these parts of the process and gives you feedback while you practise them.
How many hours a week does the course require, and can I study while working full-time?
How many hours a week does the course require, and can I study while working full-time?
That depends on how you're funding it. If you're paying for the course yourself, it's 12 hours a week, over 3.5 to 4.5 months, designed to fit around full-time work. If you're studying through the Germany-funded Bildungsgutschein route, it's 30 hours a week of study time, over 3 months, treated as full-time and not compatible with a full-time job.
You might also see "40 academic hours" mentioned for the Germany-funded route. That's the same 30 hours, just measured in 45-minute academic-hour units rather than 60-minute clock hours, which is how German funding paperwork counts study time. Same commitment, different unit.
Can a Bildungsgutschein cover this course?
Can a Bildungsgutschein cover this course?
Eligible German residents can study the approved three-month programme with a Bildungsgutschein, which covers the tuition cost. The Agentur für Arbeit decides eligibility after an individual consultation, so German residence alone does not guarantee funding.
What certificate will I receive?
What certificate will I receive?
You will receive one Turing College certificate for Building with AI Agents after completing the required core programme. The global certificate records 200 hours. The Germany-funded certificate records 520 academic hours (Unterrichtseinheiten). The certificate lists the programme parts you have completed when it is issued. The global capstone is not required for the core certificate.
What happens if I get stuck?
What happens if I get stuck?
You can attend weekly stand-ups, join Open Sessions shared with AI Engineering learners, and get one-to-one project reviews with experienced professionals to troubleshoot problems and receive feedback on decisions.
Is the Turing College Building with AI Agents programme right for you?
Is the Turing College Building with AI Agents programme right for you?
We would rather you choose the right programme than enrol in the wrong one.
This programme is for you if:
- ✅ You want to learn how to turn ideas, products or manual processes into working software.
- ✅ You want to build full-stack web applications, mobile apps, AI assistants or SaaS products by directing AI coding agents.
- ✅ You do not have a coding background, but you are comfortable learning digital tools, checking results and solving problems when an application does not work as expected.
- ✅ You want a repeatable process for planning, building, testing and deploying software, rather than relying on trial-and-error prompting.
- ✅ You can commit around 12 focused hours per week for 3.5 to 4.5 months.
- ✅ You live in Germany and can study for 40 academic hours per week over 13 weeks through an approved funded route.
This programme is not for you if:
- ❌ You mainly want to use AI for analysis, content and workplace automation without building software. AI for Business is a better fit.
- ❌ You already have at least one year of Python or JavaScript experience and want to specialise in LangChain, RAG and multi-agent systems. AI Engineering goes deeper into building LLM applications in code.
- ❌ You want to train machine-learning models or study statistics and model theory. Data Science and AI is designed for that path.
- ❌ You want a passive, video-only course. This course requires you to build projects, meet deadlines and respond to feedback from Senior Team Leads.
- ❌ You need a university-accredited degree. This is a professional course that awards a Turing College certificate.
- ❌ You expect an AI agent to produce reliable software without human review. You will need to test its work, manage access and make the final decisions.
What is the difference between Building with AI Agents, AI Engineering and AI for Business?
What is the difference between Building with AI Agents, AI Engineering and AI for Business?
The main difference is what you want to create and how much coding experience you already have. Building with AI Agents teaches you to direct coding agents to create software. AI Engineering teaches experienced programmers to build LLM applications in code. AI for Business focuses on using AI and automation.
| Choose this programme | If your main goal is |
|---|---|
| Building with AI Agents | Direct AI coding agents to build and deploy web, mobile and AI-powered applications without needing previous coding experience. |
| AI Engineering | Use existing Python or JavaScript skills to build LLM applications, RAG systems and multi-agent systems in code. |
| AI for Business | Use AI for workplace analysis, communication, decision-making and automation without becoming a software builder. |
Next group starts on November 9th