pytest Experts in Germany

AI Readiness - Part 3: Creating an e-book on the use of AI in software testing

• Designed and created an e-book on the sensible and safe use of AI in software testing from the perspective of an experienced test expert.
• Used generative AI (ChatGPT, Perplexity Pro) as a sparring partner for structuring, wording suggestions and creating variations, with all content finally reviewed and approved using my own expertise.
• Developed practical insights into the opportunities and risks of AI in testing, with a special focus on collaboration between testers and AI instead of “tester vs. AI”.
• Title: “11 starke Gründe: Dream-Team statt Rivalen - Tester und KI sind gemeinsam unschlagbar”
• Subtitle: “Wie Mensch und Maschine im Softwaretest das perfekte Team bilden – und warum reines KI-Testing riskant ist.”
• Planned use of the e-book as a basis for workshops, customer meetings or internal training.
• Technology: Generative AI, ChatGPT, Perplexity Pro, Microsoft Word, Adobe Acrobat, many years of hands-on professional testing experience, common sense.

• Optimized the data architecture (Medallion) to better decouple processing steps and improve transparency and reproducibility

• Ensured the technical quality of data processing in Databricks by introducing schema enforcement, data quality checks, and a structured data architecture

• Pipeline orchestration with Azure Data Factory

• Improved and automated the development and deployment process by integrating Git and GitHub Actions

• Technical leadership of the Data Engineering team (3 employees)

• Led workshops to optimize and stabilize the data platform and development process

• Collected and prioritized new requests and maintained the product backlog

• Technologies: Microsoft Azure (Data Lake, Data Factory), Databricks, Apache Spark (PySpark), Python, SQL, Git, Confluence, Power BI, Power Apps, Dataverse, MS Sharepoint, Mural

• Claude Code subagents, MCP, Pydantic V2, pytest, bandit

• Designed and shipped 3 specialized agents that hand work down a line: a research agent writes a cited implementation spec, a coding agent builds the modular code and its tests, a review agent ranks findings by severity and applies the fixes. Each handoff is a structured document, so no stage depends on another agent's context window.

• Connected the research agent to an academic-research MCP server (Semantic Scholar, ArXiv, Hugging Face Hub, citation snowballing) so every reference traces to a tool result rather than the model. Gated commits behind ruff, mypy, pytest and bandit, required human sign-off before installs and commits, and persisted session state on disk so long runs survive a context reset.

Project: Defect Generation on Test-Bench Images of Metal Surfaces Environment: Automated Visual Inspection (AVI), Metallurgy & Manufacturing

• Objective & Implementation: Designed, architected, and trained Generative Adversarial Networks (Pix2PixHD / SPADE) for image-to-image transformation. Targeted generation of synthetic material defects (e.g., cracks, inclusions, scale) on rough metal surfaces under real test-bench lighting conditions for privacy-compliant and efficient dataset expansion (data augmentation).
• Technical Design: Implemented robust Generative AI and computer vision pipelines in Python and PyTorch. Used semantic segmentation approaches for mask-controlled defect synthesis and subsequent evaluation with EfficientDet object detection models.
• Business Impact: Massive dataset upscaling (10x) without time-consuming and costly physical test-bench runs, while significantly improving the detection performance of automated inspection systems.

Technologies & Skills Used: Python | PyTorch | SPADE | Pix2PixHD | EfficientDet | Machine Learning | Semantic Segmentation | Computer Vision

• Technical leadership across hardware, firmware, and software teams; scope: hardware/firmware team (4 people) and leadership group (5 people)
• Consolidated and documented a product family that had grown over more than 15 years and aligned it with CRA compliance — from the bare-metal I/O module to the cloud interface.
• Provided the most important customer product with the essential requirements and architecture documentation within two months — for a firmware landscape that had grown over more than 15 years. It now supports the customer’s modernization strategy.
• Established a monthly reporting line to the supervisory board and executive board within three months: nine meetings since 12/2025. The report itself is versioned and built from the CI pipeline; it is based on automatically collected activity and release data instead of assessments.
• Built a container-based CI/CD infrastructure from scratch: cross-compilation, host tests, and documentation builds in one continuous pipeline.
• Introduced declarative QA gates for DevOps and development artifacts — from the start using lefthook instead of pre-commit, executed in a dedicated container image.

Technologies used: arc42, req42, tpo42, docToolchain, PlantUML, ArchiMate, C4 model, ADR, C, C++ (GTest), CMake, Bare Metal (ARM Cortex-M3/M7), OCI containers, Jenkins, lefthook, Prometheus, Grafana, SBOM, CRA, OPC, SCADA, PLC integration, IPv6 migration, Zero Trust, Sociocracy 3.0, Cynefin

Temporary employment

• System black-box integration tests (BBIT, IVVQ): Execution of regression, release, confirmation and conformity tests for safety-critical braking control units in the railway industry
• Software Test Application & Integration: Runtime configuration of software components and libraries, validation of interfaces, configuration dependencies and component interactions
• Test Automation (FEAT Framework): Co-development and further development of an automated test framework for test execution, reporting and result analysis
• Functional Safety (SiL4, FuSi): Ensuring compliance with safety requirements, traceability and coverage as well as conformity with EN50126/28/29
• Test Automation for Communication Components: Configuration and validation of fieldbus (CAN) and Ethernet-based TCMS data communication interfaces (TRDP and CIP)
• Requirements Analysis & Shift-Left (PTC Windchill ALM): Analysis of software and system artifacts to identify gaps, ambiguities and redundancies early in the SDLC
• Test Design & Test Case Development: Derivation of test conditions and coverage strategies and implementation of data-driven test cases (DDT) including reusable test data fixtures
• CI/CD & Automation (Python, PowerShell, Jenkins, SVN): Automation of build, test and HIL deployment processes as well as integration into CI/CD pipelines
• Test Data & Configuration Management (XML): Maintenance and adjustment of XML test vectors and system configurations with automated integration into test environments
• Non-Functional Testing: Execution of performance and load tests to assess stability and system behavior
• Agile Development & Defect Management (JIRA, Confluence): Work in Scrum teams, test coordination, review of test artifacts as well as defect tracking and root-cause analysis
• Error Analysis & Debugging (CANoe, CANalyzer): Analysis of errors and message flows across multiple system levels (application to bus)
• Model-Based Analysis (UML, Enterprise Architect): Specification of SUT/SOW and support for systematic test control
• Process & Test Documentation: Creation of integration and test documentation according to internal quality and certification requirements

Multi-Agent Document-Generation Platform

Problem:

Turn a single prompt into a well-researched, structured and cited document, while keeping the user in control: every stage can be reviewed, corrected or revisited before the system moves on.

Architecture:

A multi-agent system orchestrated by a central controller. More than 20 specialised pydantic-ai subagents (understanding, structuring, planning, research, writing, assembly, cross-reference check, glossary, citation, reference table and more) pass results forward through a shared plan persisted in Postgres (asyncpg); a runner layer dispatches each task by name to its agent, and mechanical steps such as formatting, table of contents and list of figures are deterministic code instead of LLM calls. Provider (Anthropic or OpenAI) is a configuration setting, and responses stream over Server-Sent Events.

Hosting & Infrastructure:

Stateless container run as a Scaleway Serverless Container on a private network with managed PostgreSQL, object storage for finished documents and assets, and centralized log shipping (Loki / Scaleway Cockpit), provisioned with OpenTofu modules and deployed through GitLab CI; usage is metered per API key.

My Contribution & Key Decisions:

• Central decision controller: a satisfaction classifier turns the user's free-text feedback into a verdict (keep working, advance, cancel or revise an earlier task), and a middleware applies it to the plan. Plan progression supports jumping back to earlier tasks and re-running the stages after them.
• Research and writing fan-out: planning produces a plan from which a research-design agent generates individual research tasks and a writing-design agent generates individual writing tasks, each linked to the research it needs. Every task runs in its own dedicated agent with its own conversation history.
• Self-review and human-in-the-loop: after each step a critical-thinking agent reviews the output; the user can accept, refine, or ask questions about a critique before the plan advances.
• MCP integrations: two remote MCP servers, web search and internal document search, connected through pydantic-ai with retries and long-running tool-call timeouts. Classifier agents decide per task whether web search is needed, and a small model condenses raw search results into structured findings before they reach the writing agents.

Usage Metering:

Clients authenticate with an API key validated against a key-management service; consumption is charged in credits calculated from token usage per model plus a markup, and exhausted quota is answered with HTTP 429.

• Development of an AI-powered project organization tool for a civil engineering company that intelligently links project, task, tender, schedule, and document data through a knowledge graph.
• Implementation of AI features for document analysis, information extraction, context-based assistance, and voice-based data capture based on Microsoft Azure AI, reducing administrative effort, making information available faster, and supporting project teams in decision-making.
• Tech stack: Python, React, TypeScript, FastAPI, Claude Code, Codex, Graphify, PostgreSQL, Microsoft Azure AI Foundry, Azure OpenAI, Azure AI Speech, Azure AI Document Intelligence, Microsoft Graph, Microsoft Entra ID, Docker, Git, CI/CD.

• Architected and deployed end-to-end machine learning pipelines across classification and prediction datasets, ensuring robustness and reproducibility through MLOps best practices.
• Contributed directly to LLM model output accuracy improvement by designing and engineering specialised prompts grounded in end-to-end ML and SciML pipeline logic.
• Developed training data for large language models by formulating coding problems that models could not resolve and subsequently documenting the correct solutions.

• AI Integration (LLM & RAG): Designed and built an internal intelligent RAG system (Retrieval-Augmented Generation) based on LLMs, n8n, and vector data to automatically analyze technical documents and error logs.
• Design & Implementation: Designed a robust RS-232/UART communication interface for an SBC-based embedded device used to control medical shaker systems.
• Architecture & Protocol Design: Built a highly maintainable software structure (OOP, SOLID) and defined resilient, hardware-level communication protocols, including multithreading and advanced error handling.
• Quality Assurance & DevOps: Automated testing with xUnit, ran integration tests directly on the hardware target, and maintained technical documentation to strict medical technology standards via Azure DevOps.

Label: C#, .NET, LLMs, RAG, n8n, RS-232, UART, Multithreading, async/await, xUnit, gRPC/protobuf, Blazor, MudBlazor, EF Core, Visual Studio 2026, Azure DevOps

• Led end-to-end delivery of ML/AI applications for the QuantistryLab simulation platform, from requirements definition and solution design to development, testing, deployment, and production support, coordinating efforts across back-end, DevOps, and ML teams
• Designed and implemented a compound search system with an open-weights LLM, the Pydantic AI agent framework, and the Model Context Protocol (MCP), exposing it through FastAPI endpoints
• Deployed and configured vLLM-based LLM serving on Kubernetes
• Built CI/CD pipelines and Helm charts to automate deployments across staging and production environments
• Instrumented the search system with OpenTelemetry spans and traces, integrating with the Honeycomb observability platform to monitor component-level latency and functional errors
• Developed a custom evaluation dataset and implemented evaluations for retrieval relevance, answer correctness, tool-call accuracy, and end-to-end task success
• Reduced infrared spectrum prediction time by 83% (from 60 to 10 minutes) by replacing a quantum-mechanical model with a pretrained graph neural network; fine-tuned the model using PyTorch, built and containerized the inference pipeline with Docker, and integrated it into the platform
• Implemented an MLflow model registry with MinIO-backed artifact storage
• Advised sales and science teams on ML/AI applications for prospective client projects, providing technical expertise to shape proposed solutions and prototypes

Technologies: AWS, Hetzner, Kubernetes, Helm, Docker, vLLM, Pydantic AI, MCP, Qwen, Llama 3, MLflow, MinIO, GitHub Actions, Python, FastAPI, OpenTelemetry, Honeycomb, PyTorch, Polars, pytest

• Developing a model-based AI where natural language is interpreted as formal relations.
• Natural language terms are not considered rigid but fluid and can be negotiated in a context so meaning resolves by iteratively specifying.
• Develops all kinds of model solutions.
• Backed by natural language and data annotation.
• Requirements to code and other solutions.

• Designed and developed a multi-tenant SaaS platform enabling organizations to build their own knowledge bases and chat with brand-customized AI assistants (white-label approach with dynamic branding per organization).
• Implemented a scalable RAG architecture with a GPT-4o tool-use loop, hybrid semantic search, and strict tenant isolation at database and search index level.
• Built persistent, project-like chat sessions including a streaming API (SSE), multilingual support, and speech input/output (STT/TTS).
• Delivered the cloud infrastructure as Infrastructure-as-Code, fully automated per-customer CI/CD pipelines, and an onboarding process for new tenants.

Technologies used: Python, FastAPI, Pydantic (v2 noted), Next.js, React, TypeScript, Tailwind CSS, OpenAI / LLMs (GPT-4o), Azure AI Search, Cosmos DB, Azure Blob Storage, Azure Cognitive Services Speech, Azure App Service, Azure Container Registry, Retrieval-Augmented Generation (RAG), Server-Sent Events (SSE), Docker, Terraform, GitHub Actions, REST, OpenID Connect (OIDC), Multi-Tenancy

Designed and developed an Azure-based AI and data platform transforming distributed project documents into searchable, structured knowledge for an enterprise AI assistant. Combined incremental document processing, hybrid keyword/vector retrieval, knowledge graph extraction, and conversational AI for Microsoft Teams and Microsoft 365 Copilot. Implemented Python document-processing pipelines for PDF, Word, Outlook and text files; Azure Cosmos DB lifecycle management; Azure AI Search; Azure AI Foundry and Azure OpenAI; identity-aware retrieval with Microsoft Entra ID; project authorization; a TypeScript Microsoft Agents SDK agent; Pulumi infrastructure; resilient scheduling, retry and diagnostic tooling. Technology stack: Python, TypeScript, FastAPI, Pydantic, Azure AI Foundry, Azure OpenAI, Azure AI Search, Azure Cosmos DB, Blob/Table Storage, Container Apps/Jobs, Service Bus, Key Vault, Entra ID, Docker, Pulumi, pytest, Microsoft Teams and Microsoft 365 Copilot.