Skip to content

Gunther Popp

Software Architect for AI‑Driven Development & Agentic AI

I am a freelance software architect and consultant with over 20 years of experience in complex projects.

I support companies in designing and modernizing long-lived software landscapes, in introducing AI‑Driven Development to their engineering teams, and in using Agentic AI in their business processes.

Gunther Popp – software architect, consultant and author

Latest posts

All posts →
  • Context EngineeringContext Management

    10/11/2026

    Best Practices: Context Engineering

    How is the context of AI coding agents created, what mechanisms manage it automatically, and how can it be influenced?

    Read more →
  • Feature SpecificationFeature File

    9/17/2026

    Best Practices: Feature Specification

    Feature specifications form the foundation of AI-Driven Development. This article explains what they should contain, how they differ from data models and implementation plans, and how to maintain them as Documentation as Code.

    Read more →
  • Claude CodeOpenAI CodexCursor & Grok

    8/6/2026

    Claude Code vs. Codex vs. Cursor

    Three AI coding agents, one specification, the same website: Claude Code with Opus 5, OpenAI Codex with GPT-5.6 Sol, and Cursor with Grok 4.5 in a direct comparison.

    Read more →

Services

My main areas of work, from strategic planning through technical implementation.

AI‑Driven Development

Design, validation, and adoption of sustainable AI‑Driven Development processes

  • End-to-end process design from specification through review
  • Pilots and step-by-step adoption in teams
  • Hands-on work with Codex, Claude Code, Cursor, and development environments such as Visual Studio Code or IntelliJ
  • Governance, quality criteria, and continuous improvement

Agentic AI for business processes

Design and implementation of AI agents that support business processes with tools and business context

  • Architecture of agent platforms and agents for business processes
  • Development of MCP servers (Model Context Protocol) as tools for agents
  • Business objects and domain knowledge as context for agents
  • Implementation on Amazon Bedrock

Architecture, APIs & backend (Java/Spring)

Pragmatic, robust architectures based on ADRs, reference architectures, and APIs

  • API design & API conventions
  • Architecture blueprints, reference architectures & ADRs
  • Microservices with Spring Boot and REST
  • Domain & interface modeling

AWS & events

Pragmatic use of AWS for messaging and integration

  • Amazon SNS & SQS integration
  • API‑Gateway/ALB connectivity
  • IAM concepts & policies
  • Cost and operational aspects

AI‑Driven Development

I design, validate, and establish AI‑Driven Development processes that connect requirements, architecture, implementation, reviews, and tests end to end. Adoption is iterative, measurable, and backed by hands-on use of suitable AI agents.

End-to-end process

  • From requirements and architecture through implementation, tests, and review
  • Clear roles, approvals, and responsibilities for people and agents
  • Tools selected for the project context, risks, and security requirements

Context & documentation

  • Requirements, architecture, and a semantic data model as the shared foundation
  • Bootstrap prompts and always-on instructions for stable context
  • Versioned implementation plans and verifiable results

Pilot & validate

  • Representative pilot projects with defined quality and efficiency criteria
  • Validate process and tools in the real project context
  • Embed findings step by step in teams, standards, and governance

From design to adoption

  • Analyze the current situation, goals, development process, and organizational constraints
  • Design the target process for specification, architecture, planning, implementation, tests, and reviews
  • Select and practically evaluate suitable agents, IDEs, and integrations
  • Implement pilots hands-on with Codex, Claude Code, Cursor
  • Validate quality, productivity, risks, and adoption against agreed criteria
  • Enable teams and embed the process step by step in standards and working practices
  • Feed experience back into templates, prompts, and guardrails

Governance & quality

  • Responsibilities, approval points, and handling of agent-generated code
  • Automated tests, architecture checks, and AI‑assisted reviews
  • Static code-security analysis, dependency and license checks, and policy checks in CI/CD

Profile and experience

For more than 20 years, I have supported clients in implementing, migrating, and renewing complex software systems. This includes designing target architectures and API standards and, more recently, introducing AI‑Driven Development processes. I also support practical implementation in teams with Codex, Claude Code, Cursor, and development environments such as Visual Studio Code or IntelliJ.

Alongside AI‑Driven Development, my second focus is the design and implementation of Agentic AI for business processes. In this area, I support the modeling and provision of domain knowledge for the agents as well as the concrete implementation of tools such as MCP servers.

Detailed profile and project history →

Contact

For professional exchange or project enquiries, contact me by email or phone.

Contact details

gpopp@guntherpopp.de

+49 160 820 89 13

Feldkirchen (Munich)

Experience and working style

  • 20+ years of software engineering experience
  • Focus areas: AI‑Driven Development and Agentic AI
  • Pragmatic and outcome‑focused approach
  • Flexible collaboration and transparent communication