Training course · AI Agent Development · Software Architecture

AI Agent Development for Software Architects

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

“AI Agent Development for Software Architects” is designed for roles such as Software architects and solution architects, Enterprise architects with links to application development and Senior developers and tech leads with architectural responsibility and shows step by step how to use AI Agent Development productively, safely and transparently in Software Architecture – from the fundamentals to advanced workflows.

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.

When AI agents write large parts of the code, architecture becomes more important, not less: clear module boundaries, documented decisions and verifiable quality goals are the frame that keeps generated code consistent. AI assistants also support architects directly, for example by analysing existing codebases, producing C4 diagrams, drafting architecture decision records and weighing alternatives against quality attributes. Fitness functions and architecture tests can be generated and anchored in your pipelines. In this course you learn to use AI as a sparring partner for architectural decisions, prepare architectural knowledge so that coding agents follow it reliably and spot erosion early in fast-growing codebases.

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?

  • Software architects and solution architects
  • Enterprise architects with links to application development
  • Senior developers and tech leads with architectural responsibility
  • Platform and cloud architects
  • Owners of architecture governance and review boards

Typical challenges we address

  • Architecture documentation goes stale quickly and drifts from the actual code
  • Decisions are made verbally and their context is lost
  • Gradual architectural erosion through cyclic dependencies and violated layer boundaries
  • Without clear guidance, AI-generated code ignores established patterns and conventions

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.
  • Use AI to analyse existing systems, evaluate architectural options and record decisions as ADRs.
  • Anchor architecture rules as fitness functions in the pipeline and detect erosion early.
  • Shape context and rules so that AI coding agents reliably respect your architectural guidelines.

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: Software Architecture
  • Documenting architecture decisions
  • Reconstructing existing architecture
  • Evaluating alternatives
  • Generating fitness functions
  • Guardrails for coding agents
  • Carving out domains
Module 7 Data protection, quality & transfer to daily work
  • Architecture documents often describe security-relevant details such as network zones and data flows: you learn which information may be shared with AI services, how to assess licensing risks of components and how architectural decisions traceably reflect GDPR, NIS2 and Cyber Resilience Act requirements.
  • 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

Documenting architecture decisions

Turn discussions, meeting notes and pull requests into structured architecture decision records with context, options considered, decision and consequences, stored consistently in the repository. Related decisions are linked automatically.

Reconstructing existing architecture

Analyse existing codebases, extract modules, dependencies and interfaces and generate up-to-date C4 diagrams as diagrams-as-code with PlantUML, Mermaid or Structurizr. Diagrams can be refreshed automatically on every build.

Evaluating alternatives

Compare options such as a modular monolith, microservices or event-driven integration against quality scenarios, cost and team structure, and make trade-offs visible following ATAM principles.

Generating fitness functions

Express architecture rules such as layer access, forbidden dependencies or naming conventions as automated tests with ArchUnit, dependency-cruiser or similar tools and wire them into CI.

Guardrails for coding agents

Describe architectural principles, module boundaries and patterns in rule files and context documents so AI agents follow them reliably when generating code, and spot deviations in review.

Carving out domains

Derive bounded contexts, aggregates and context maps following domain-driven design from event storming results and business descriptions, and use them to shape teams and services.

Possible hands-on projects

ADR assistant

A command-line tool that turns bullet points into a complete ADR from a template, links related decisions and updates an index in the repository.

Dependency visualiser

A script that analyses import relationships in a codebase, flags cycles and layer violations and renders an interactive diagram in the browser.

Architecture Q&A over repos

A small application that indexes ADRs, diagrams and code and answers architecture questions with references to the underlying sources. New team members onboard faster.

Prerequisites

Solid programming skills in Python or TypeScript, experience with REST APIs and some hands-on experience with LLM APIs or AI assistants. You have several years of development experience and have made or supported architectural decisions in larger software systems.

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

Architecture documents often describe security-relevant details such as network zones and data flows: you learn which information may be shared with AI services, how to assess licensing risks of components and how architectural decisions traceably reflect GDPR, NIS2 and Cyber Resilience Act requirements.

Suggested dates

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

  • Tue, 24 November 2026 – Wed, 25 November 2026 Online Request
  • Thu, 3 December 2026 – Fri, 4 December 2026 Hanover Request
  • Thu, 17 December 2026 – Fri, 18 December 2026 Online Request
  • Thu, 7 January 2027 – Fri, 8 January 2027 Online Request
  • Thu, 21 January 2027 – Fri, 22 January 2027 Munich Request
  • Tue, 2 February 2027 – Wed, 3 February 2027 Online 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.

Can AI make architecture decisions?

AI is a strong sparring partner: it offers options, names trade-offs, finds gaps in reasoning and documents decisions in a structured way. Responsibility stays with people, because architecture depends on business context, team skills, budget and risk appetite. In the course you learn to use AI so that decisions are better reasoned and made faster.

How do we stop AI agents from eroding the architecture?

Through a combination of clear context and automated checks: rule files and ADRs describe module boundaries and patterns, fitness functions and architecture tests in CI block violations, and reviews focus on structural changes. Architecture thus turns from documentation into a verifiable guardrail for humans and agents alike.

How much does “AI Agent Development for Software Architects” 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 Software Architects

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