Agentic AI Cohort — Build an AI-Powered App
Master agentic workflows by shipping a real app
Build "Expense AI Agent" — an AI-powered agent application with function calling, a Telegram bot, a web dashboard, and full deployment. 6 weeks, hands-on, project-based.
Price: €2,000 · Duration: 6 weeks · Format: Group cohort · Next start: April 20, 2026
Who is this for?
Intermediate Python developers who want to build AI applications with real engineering — not prompt wrappers or API demos. You'll learn structured outputs, function calling, prompt engineering, and how to architect testable, deployable AI applications with 95%+ test coverage. The same code review and shipping discipline applied across every program.
Time commitment: ~10 hours per week.
Program overview
What you build, week by week
Scaffolding
- Environment setup
- Repository pattern
- SQLModel entities
- Python StrEnum
- In-memory data layer
Assistant Setup
- Python Protocols
- Type aliases
- Pydantic structured outputs
- OpenAI function calling
- LLM client abstraction
Tools for the Agent
- Prompt engineering
- Service layer pattern
- CLI with Typer + Rich
- Database repositories
- Classification service
Telegram Integration
- Input preprocessing
- Conversation handlers
- Human-in-the-loop (HITL)
- Inline keyboards
- Telegram bot interface
Web Interface
- FastAPI fundamentals
- Pydantic schemas
- Streamlit dashboard
- API integration
- REST API + Web UI
Deploy + Docs
- Docker fundamentals
- CI/CD pipelines
- Test coverage (95%+)
- Documentation
- Production-ready agent
Tech stack: SQLModel, Pydantic, OpenAI function calling, Typer + Rich, FastAPI, Streamlit, Telegram bot API, Docker, CI/CD
What's included
- Weekly group call with your coaches
- Code review via GitHub PRs — detailed, honest feedback
- Professional, portfolio-ready project on GitHub
- Structured 6-week curriculum
- Access to invite-only community
Juanjo's coaching in action
Juanjo has coached developers 1:1 on AI projects before launching this cohort:
"In 6 weeks I built a full RAG pipeline — PDF ingestion, vector embeddings, and an LLM query layer that returns cited answers. The most surprising thing? It's not a huge technical challenge in Python. Even if you consider yourself a beginner, it's completely within reach."
— Tim G. · Built AI-powered search for QuietLinks
Your coaches
Juan José Expósito González — Python & AI Mentor and PhD Engineer. Guides developers from Python basics to advanced AI implementations. Expert in Python, machine learning, blockchain, and algorithmic trading. Passionate about transforming complex concepts into practical, clean architecture, testable and deployable solutions.
Bob Belderbos — Developer coach and builder with 11 years at Sun/Oracle and 6+ years running Python coaching programs. Co-founded Pybites and built Pybites Platform (400+ Python exercises). 100+ developers coached. Bob brings the architecture perspective — clean code, testable design, and bridging the gap between AI prototypes and production-ready applications.
Frequently asked questions
What community support is there? You get access to our invite-only community where you can ask questions, share progress, and connect with other developers.
Do I need AI/ML experience? No prior AI experience needed. You should be comfortable with Python — intermediate level. We teach the AI patterns through the project.
What if I fall behind? Sessions are recorded. You have async support from Juanjo and Bob throughout the week.
Do I need an OpenAI API key? Yes, you'll use the OpenAI API for function calling and structured outputs. Cost is minimal (a few dollars for the full program).
Will this help my portfolio? Yes — an AI-powered app with agent tools, a Telegram bot, a web dashboard, and 95%+ test coverage is a strong portfolio piece.
How do I get started? Book a free call → and we'll talk about whether this is the right fit.
In 6 weeks, you'll have a deployed AI agent app. Not a tutorial, a real product.