There is no upcoming events yet. Sign up to our newsletter if you want to get notified whenever new events get announced.
Full-Time CourseAI Engineering Program
Become a production-ready AI Engineer in 9 weeks. Build, evaluate and deploy LLM applications with RAG, AI agents and MCP, from prompt engineering to production on AWS.
Full-Time
weeks
Munich
English
Program overview
Start with language models and prompting, publish your first AI demo in week two, then turn it into a real service and grow it every week: evaluation, agents, security and production operations, the same system carried all the way through. By the end, you will be able to take on a real production AI system with confidence, not just recite a list of tools.
Awards
Upcoming dates
The next date is not yet known
Are you interested in applying for our course? Then register here and we will inform you as soon as the next course dates are published.
Looking for financing? Check out our financing options.
Schedule
- Mon - Fri
On-site
- 09.00 - 12.00Lecture
- 13.00 - 18.00Practice
LectureLearn from our instructors who are experts in their respective fields and get introduced to new topics during live lectures.
PracticeWork on a set of interesting and challenging exercises related to the topics covered during morning lecture. Practice your team-building skills by doing group projects together with your peers.
Where our students get jobs
Testimonials
Till Solenthaler
Full-Stack
I loved their hands-on approach and they always taught us best-practices with real-life implementations in mind.
Went from System Engineer to Software Engineer at SIX
Leal Oburoglu
Mastering Generative AI
I definitely recommend this course. I think everyone, at any career stage, should take it. The sooner we learn about generative AI, the better, as it will continue to expand rapidly!
Went from Associate Researcher to Startup Founder at Sci.Alpha
Mijail Febres
Full-Stack
Constructor Academy supported me effectively with assistance when preparing for my interviews.
Went from Scientist to Software Developer at Repower
What you will learn
Week before start
Open session
Meet your fellow students for an evening session the week before the program starts. Set up your laptop and exchange questions with the class.Week 1
Language models & prompting
- LLM Fundamentals: Scope an AI product and learn tokens, transformers and sampling.
- Multi-Vendor APIs: Call OpenAI, Anthropic and Gemini from Jupyter, coding faster with Claude Code and Cursor.
- Hugging Face and Ollama: Run open-source models, compare cost and quality, and trace every call with Langfuse.
- Prompt Engineering: Design reliable, versioned prompts with patterns that work.
- Structured Outputs and Gradio: Get validated data with Pydantic and build a streaming chat interface.
Week 2
Retrieval-Augmented Generation (RAG)
- Document Ingestion: Parse PDFs, tables, scans, web pages and transcripts.
- Chunking and Embeddings: Split documents without losing context and search them by meaning.
- RAG Pipeline: Generate grounded answers with citations, measured with recall and MRR.
- Hybrid Search: Combine BM25 and vector search, then rerank to beat your baseline.
- Hugging Face Spaces: Publish your first RAG demo and share the link.
Week 3
AI backend engineering
- FastAPI Backend: Grow your RAG prototype into an async, streaming API traced end to end.
- PostgreSQL and pgvector: Move RAG search into your database, with an identity for every user.
- Reliable AI Endpoints: Survive provider outages, validate model outputs and test AI code with pytest.
- Function Calling: Let the model call your backend's functions through a hand-built agent loop.
- Langfuse and Simulated Traffic: Debug with traces, version prompts and simulate realistic users.
Week 4
Evaluation (Evals)
- Error Analysis: Read real traces systematically to find what actually goes wrong.
- Golden Datasets: Build test sets with known answers, including synthetic data.
- LLM-as-a-Judge: Grade faithfulness, relevance and correctness, aligned with human labels.
- Eval Gates: Choose prompts and models with data, and block regressions in CI.
- Cost Optimization: Cut token cost with Redis caching and model routing.
Week 5
AI agents
- LangGraph: Rebuild your agent loop with state, checkpoints and clear control flow.
- Agentic RAG: Let agents decide when and how to search, and compare LangChain and Pydantic AI.
- MCP: Build MCP servers and connect tools and skills through the Model Context Protocol.
- Memory and Agent Evals: Give agents per-user memory and evaluate every step they take.
- Coding Agent: Build a sandboxed agent that reads, edits and tests code.
Week 6
Multi-agent systems & security
- Multi-Agent Systems: Orchestrate agents and subagents, including multimodal and computer-use agents.
- Human-in-the-Loop and Guardrails: Require approval for risky actions and block unsafe inputs and outputs.
- Prompt Injection and Red-Teaming: Attack your own system using the OWASP Top 10 for LLMs.
- GDPR and EU AI Act: Protect personal data, meet AI-literacy requirements and test security on every release.
- Team Hackathon: Build a deep-research or multi-agent system with your team in one day.
Week 7
Deployment & monitoring
- Docker Hardened Images: Secure your stack with minimal, non-root images and supply-chain scanning.
- AWS Foundations: Deploy to EC2 with IAM, S3, Secrets Manager, Route 53 and HTTPS.
- Data Safety and Multi-User Isolation: Back up and restore your pgvector database, isolate each user's data and release with rollback.
- Prometheus and Grafana: Monitor quality, drift and cost, and load-test your system.
- Ollama and vLLM: Self-host open-source LLMs on an AWS GPU and fine-tune embeddings with LoRA.
Week 8 – 9
Capstone project
- Your Own AI System: Combine RAG, agents and MCP tools into one production system for a new domain.
- Scoping & Design: Turn a stakeholder brief into a system design and defend it in an architecture review.
- Build & Deploy: Develop your system and ship it live on AWS with monitoring.
- Evaluation & Security Review: Prove its quality with evals and test it against attacks before release.
- Demo Day: Polish your portfolio and present your live system publicly.
Mentorship
At Constructor Nexademy, we mentor our students, with a focus on placing their individual needs and goals at the center of our approach. Our goal is to empower our students to succeed by providing them with the guidance and support they need to achieve their full potential.
Ongoing mentorship
No need to schedule appointments; receive prompt and continuous feedback. Our teaching assistants are readily available to assist you.
Real-world projects
Effective mentoring equips you with the skills to tackle actual work challenges. Our capstone projects mirror real industry projects, bringing together all that you have learned.
Career coaching
We assist you in finding new job opportunities and showcasing your qualifications to potential employers.
Live lectures
Learning can be tough, and that's why the dropout rate for self-paced courses is as high as 85%. We recognize that interactive, human-led instruction is crucial to achieving ambitious learning objectives.
Topics
Language Models & Retrieval
Work with language models from OpenAI, Anthropic and Hugging Face, apply prompt engineering, and code productively with AI assistants. Build a full RAG pipeline over real-world documents, with hybrid search and citations that reduce hallucinations.
Production Services & Evaluation
Grow your RAG prototype into a traced FastAPI service backed by PostgreSQL. Evaluate answer quality with LLM-as-a-judge and RAG evaluation, and wire eval gates into your CI pipeline so a bad release never ships.
AI Engineering
This program starts with language models and prompting in week one, then builds one production AI system from week two onward rather than teaching a checklist of tools. You will master retrieval, evaluation, single- and multi-agent systems, and production operations. By the end, you will have shipped, evaluated, secured and operated a production AI system, with code review from a practising AI engineer along the way.
- Production LLM services, retrieval, evaluation, agents, and cloud deployment
- A strong foundation for AI engineering roles that build and run LLM products in production.
- Ideal for software engineers with prior programming experience and basic Docker knowledge who want to build production-grade AI systems
- Hands-on learning through one growing system, redeployed at the end of every week
360 hrs
hands on learning
Ship, evaluate and operate a production AI system end to end
We support you in finding your next dream job:
- One-to-one sessions with career advisors
- Cover Letter and CV writing sessions
- Sending your CV to our network of hiring companies
- In-house events such as our Hiring Day
- Opportunity to collaborate with companies on a project
Choose your location
Visit our campus in Munich
Would you like to see what your time at Constructor Nexademy could be like and where our students spend most of their time? Then contact us for a visit of our campus.
Constructor Nexademy
Landsberger Strasse 110
80339 München
Financing options
At Constructor Nexademy, we believe that finances should never be a barrier to accessing the education and training that can help individuals achieve their goals. That's why we offer a variety of financing options to make our courses more accessible to a diverse range of students. We also work with external organizations that provide financial assistance to those in need.
Certificate from a leading AI & Data Science academy
Showcase your achievement on LinkedIn, your CV and beyond. Your Constructor Nexademy certificate tells employers you've mastered in-demand skills through hands-on, real-world projects, and helps you take the next step in your career.
Upcoming events
Attend one of our events. Discover our upcoming workshops, info sessions, final presentations and webinars on trending topics.
Read more about Constructor Nexademy in our Blog
Read the latest news about Constructor Nexademy and get informed about all things around Programming and Data Science in Switzerland and Germany.
Data Science Part-Time capstone projects batch #11
by Christina Sieber
Data Science capstone project batch #36
by Christina Sieber
Data Science capstone projects batch #35
by Christina Sieber