Training course · AI Agent Development · Python Development

AI Agent Development for Python Developers

Build AI agents professionally: architectures, tool use, agent SDKs, LangGraph, memory, evals, guardrails and deployment.

Duration
2 days (14 hrs)
Format
Online or on-site
Language
German or English
Participants
open dates or exclusive for your team
Certificate
Certificate of attendance

Overview

This course connects AI Agent Development with the everyday processes of Python Development. Participants work on realistic scenarios and leave with prompts, workflows and checklists they can use straight away.

AI agents go beyond conventional chatbots: they break tasks down, call tools and APIs, evaluate the results and keep iterating until a goal is met. In this two-day course you learn to build such systems so that they work reliably, transparently and safely in day-to-day business. You compare architectural patterns ranging from simple workflows through the agent loop to multi-agent systems, and you implement them hands-on with the Claude Agent SDK, the OpenAI Agents SDK and LangGraph.

The focus is not on a quick demo but on the path to production. You design tool interfaces and context deliberately, connect tools via the Model Context Protocol, plan short-term and long-term memory and build evaluations that surface regressions early. On top of that come guardrails against prompt injection and harmful actions, human-in-the-loop approvals, tracing, cost control and deployment. By the end you can judge when an agent is worthwhile, when a deterministic workflow is enough and which framework fits your project.

Python powers web back ends, automation, data pipelines and AI applications – often inside the same organisation, which makes codebases notoriously diverse. AI coding assistants and agents have a particularly visible impact here: they write pytest suites around your existing fixtures, add type hints until mypy passes cleanly, migrate legacy scripts to modern packaging with pyproject.toml and walk you through an unfamiliar Django or FastAPI project in minutes rather than days. The key is steering these tools with the right context: project conventions, Ruff rules, test commands and architectural decisions belong in instruction files and prompts. Once you master that, you can delegate routine work with confidence and focus on design, review and business logic.

The course is delivered live – either online in a virtual classroom or on-site in our training rooms in Munich, Salzburg, Berlin, Augsburg, Hanover, Bremen, Stuttgart and Kaiserslautern. We can also run it exclusively for your team.

Who is this course for?

  • Python back-end developers working with Django, Flask or FastAPI
  • Automation and scripting specialists in IT and business units
  • Data and ML engineers maintaining production Python code
  • Tech leads responsible for code quality and review processes
  • Developers modernising legacy Python codebases

Typical challenges we address

  • Ageing codebases with no type hints and patchy test coverage
  • Dependency sprawl across requirements.txt, Poetry, uv and virtual environments
  • Time-consuming reviews and inconsistent style despite linters and formatters
  • Onboarding into unfamiliar projects that rely on undocumented tribal knowledge

What you will be able to do

  • You choose the right pattern for a task, from a fixed workflow to an autonomous agent.
  • You design tools and context so that agents work reliably and cost-effectively.
  • You implement agents with the Claude Agent SDK, the OpenAI Agents SDK or LangGraph.
  • You set up evaluations and tracing to measure quality and detect regressions.
  • Have AI agents reliably deliver pytest suites, type hints and refactorings across your Python codebase.
  • Encode project conventions, tooling such as Ruff, mypy and uv, and test commands so the assistant follows them consistently.
  • Review generated Python code systematically for correctness, security and maintainability before it is merged.

Course outline

Day 1

Module 1 Agent architectures and design patterns
  • Workflows versus agents: prompt chaining, routing, parallelisation, orchestrator-worker
  • The agent loop: plan, act, observe, reflect
  • Multi-agent systems, subagents and handoffs between agents
  • Criteria for choosing model, framework and level of autonomy
Module 2 Tool use and context engineering
  • Function calling and structured outputs with JSON Schema
  • Designing good tools: granularity, naming, error messages, idempotency
  • Connecting and reusing tools via MCP servers
  • Managing the context window: summarisation, compaction, retrieval and prompt caching
Module 3 Frameworks in practice
  • Claude Agent SDK: agent loop, tools, subagents, permissions and hooks
  • OpenAI Agents SDK: agents, handoffs, guardrails and tracing
  • LangGraph: stateful graphs, checkpointing and human-in-the-loop
  • Side-by-side comparison on the same task and criteria for choosing a framework
Module 4 Memory, evals and observability
  • Short-term and long-term memory, storage strategies and forgetting
  • Evaluation datasets, LLM-as-a-judge and task-based success metrics
  • Tracing tool calls, token usage and latency
  • Regression testing when prompts, models or tools change

Day 2

Module 5 Guardrails, security and deployment
  • Protecting against prompt injection, data exfiltration and over-privileged tools
  • Input and output checks, approval steps and sandboxing
  • Deploying as a service, asynchronous execution, retries and cost limits
  • Governance, audit logging and requirements under GDPR and the EU AI Act
Module 6 Practice lab: Python Development
  • Generating pytest tests
  • Type hints and mypy clean-up
  • Django migrations and ORM refactoring
  • FastAPI endpoints with Pydantic
  • Modernising legacy scripts
  • Understanding and documenting a codebase
Module 7 Data protection, quality & transfer to daily work
  • Define which source code may be sent to cloud-hosted models, keep secrets from .env files strictly out of context, and check generated code for licence carry-over and newly introduced PyPI dependencies (typosquatting, supply-chain risk). For products with digital elements, obligations under the Cyber Resilience Act are becoming increasingly relevant.
  • Ground rules for safe AI use in your team
  • Checking results: facts, sources, tests and the four-eyes principle
  • Personal action plan for the next 30 days
  • Open questions, outlook and resources for further learning

Practical scenarios from your field

Generating pytest tests

The assistant inspects functions and existing fixtures, writes parametrised pytest tests covering edge cases and mocks for external services, then runs them iteratively until they pass and reach the coverage you are aiming for.

Type hints and mypy clean-up

Untyped code is annotated module by module with type hints, TypedDicts and Protocols. The agent runs mypy or Pyright, interprets the output and fixes issues step by step without changing runtime behaviour.

Django migrations and ORM refactoring

Model changes are planned together with data migrations, makemigrations conflicts are resolved and N+1 queries are fixed with select_related or prefetch_related. The AI flags risks to existing production data before you deploy.

FastAPI endpoints with Pydantic

From a functional description or an OpenAPI draft, the assistant produces Pydantic models, routers, dependency injection and validation logic, complete with error handling and matching tests using the TestClient.

Modernising legacy scripts

Old scripts become proper packages with pyproject.toml, global state is removed, argparse or Click interfaces are added and Ruff rules applied. The agent documents every change clearly in the commit history.

Understanding and documenting a codebase

The assistant explores an unfamiliar repository, produces architecture overviews, explains data flows between modules and fills in missing docstrings in Google or NumPy style so new team members become productive faster.

Possible hands-on projects

Log analysis CLI

A Typer-based command-line tool that reads application logs, clusters error patterns and outputs a concise summary as Markdown or JSON – shipped with tests and proper packaging.

FastAPI microservice with a database

A small REST service with SQLAlchemy, Alembic migrations, Pydantic validation and a Docker setup, planned, implemented and tested end to end by an AI agent.

Dependency audit script

A script that lists all project dependencies, checks for outdated or vulnerable packages against public advisory databases and produces a prioritised upgrade report.

Prerequisites

Solid programming skills in Python or TypeScript, experience with REST APIs and some hands-on experience with LLM APIs or AI assistants. Solid Python skills plus hands-on experience with Git, virtual environments and at least one framework such as Django, Flask or FastAPI.

Technical requirements

A laptop with Python or Node.js, Git and a code editor, plus API access to at least one LLM provider such as Anthropic or OpenAI. Credentials for the exercises can be provided if needed.

Data protection & compliance

Define which source code may be sent to cloud-hosted models, keep secrets from .env files strictly out of context, and check generated code for licence carry-over and newly introduced PyPI dependencies (typosquatting, supply-chain risk). For products with digital elements, obligations under the Cyber Resilience Act are becoming increasingly relevant.

Suggested dates

All dates are suggestions – your preferred date is always possible.

  • Wed, 25 November 2026 – Thu, 26 November 2026 Online Request
  • Tue, 8 December 2026 – Wed, 9 December 2026 Bremen Request
  • Thu, 7 January 2027 – Fri, 8 January 2027 Online Request
  • Tue, 19 January 2027 – Wed, 20 January 2027 Salzburg Request
  • Tue, 2 February 2027 – Wed, 3 February 2027 Online Request
  • Wed, 10 February 2027 – Thu, 11 February 2027 Kaiserslautern Request

Frequently asked questions

Which agent framework should we use?

It depends on your model strategy, the level of control you need and your team’s skills. Vendor SDKs offer a lean start with tightly integrated features, while LangGraph gives you finer control over state and flow. In the course you solve the same task with several frameworks and derive clear selection criteria for your own project.

How do you stop an agent from taking wrong or dangerous actions?

With several layers of protection: narrowly scoped tools with minimal permissions, input and output validation, approval steps for critical actions, sandboxing and continuous evaluation. Prompt injection via external content is treated as a realistic threat. You practise these measures directly on your own agent.

Does every automation need an agent?

No. Many tasks are handled more reliably, cheaply and transparently by a fixed workflow of a few LLM calls. Agents pay off when the solution path cannot be planned in advance and the system must react flexibly to intermediate results. We train exactly this trade-off using practical examples.

Does AI-assisted coding work in large Django monoliths?

Yes, provided the agent gets the right context. On the course you learn to create instruction files describing architecture rules, app structure and test commands, and to scope tasks so the assistant works on individual apps or modules. Changes stay manageable and reviewable even when the overall project is very large.

Will AI replace my code reviews?

No. AI assistants speed up writing code and can prepare reviews by pointing out missing tests or risky sections, but functional and security accountability stays with your team. We show how to combine review checklists, static analysis and tests so that generated code meets exactly the same quality bar as hand-written code.

How much does “AI Agent Development for Python Developers” cost?

The course costs a flat €949 plus VAT per participant – whether online or on-site. This includes 2 training day(s) with a live trainer, course materials, a lab environment and a certificate of attendance.

Can I suggest my own date?

Yes. Our suggested dates are just a guide. In the request form you can always describe your preferred date or period in free text – we check availability and confirm within one working day.

Is the course online or on-site?

Both are possible: live online in a virtual classroom or on-site in Munich, Salzburg, Berlin, Augsburg, Hanover, Bremen, Stuttgart or Kaiserslautern. We can also run the course exclusively for your team.

More courses for Python Developers

More AI Agent Development courses

All 30 AI Agent Development courses

AI Agent Development · MunichAI Agent Development · SalzburgAI Agent Development · BerlinAI Agent Development · AugsburgAI Agent Development · HanoverAI Agent Development · BremenAI Agent Development · StuttgartAI Agent Development · KaiserslauternAI Agent Development · Online