Refactoring Experts in Dusseldorf

• 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

• Supporting the DevOps team in Terraform-managed, multi-region AWS infrastructure to achieve environment parity.
• Developed end-to-end CI/CD pipelines using AWS CodePipeline, CodeBuild, and CodeDeploy, automating the build and deployment.
• Maintained pre- and post-deployment scripts to automate critical tasks such as database schema migrations and environment sanity checks.
• Implemented CI/CD flow specifically for hotfixes via separate Git branches, managing back-merge activities from feature branches to release branches to ensure code integrity through automated conflict resolution.
• Deployed a dedicated, lightweight sanity check application hosted cost-effectively on Azure Container Apps to run automated health and basic functional checks as a post-deployment activity triggered via pipeline.
• Investigated production incidents through code changes and AWS CloudWatch logs.
• Coordinated integration of Dynatrace APM and its APIs for monitoring purposes.
• Full stack QA strategist for a high-traffic e-commerce platform built on a layered architecture for the back-end testing of core platform services, especially the order management system in Zed and Glue layers.
• Managed automation activities, testing process, and refactoring practices.
• Responsible for framework migrations, setup, and training for new automation frameworks.
• Promoted a shift-left approach within the QA team and created the test concept.
• Participated in meetings with IT managers, business owners, product owners, and team members.
• Designed and implemented contract testing to validate API schema compatibility between the order management system and the Zed and Glue layers, reducing production-relevant breaking changes by approximately 3%.
• Led the migration to a multi-environment framework that enabled test execution across 4 country configurations from a single codebase.
• Integrated automated unit and functional tests directly into the GitLab CI/CD pipeline, reducing pipeline runtime by 32%.
• Coached and trained 5 QA engineers across Germany and Hungary in test automation, framework architecture, and best practices.
• Architected a layered backend test automation framework separating business logic, API request builders, and the database layer.
• Piloted AI-assisted testing with Playwright Agents, the Playwright MCP Server, and GitHub Copilot for automated test generation, execution, and self-healing Playwright scripts.

• Led the migration of the existing DWH, including HR data, application processes, project business and time tracking, from an MS SQL Server environment to Azure Databricks
• Refactored existing data models and ETL processes and extended them with new objects and functionalities
• Used Azure Databricks (Unity Catalog, Notebooks, Spark SQL, Databricks Dataflows) and Power BI
• Technologies: MS SQL Server 2019 (SSMS, SSIS), Power BI, Delta Lake, Azure Databricks

• Identification of requirements (legal, organizational, and technical)
• Design of solution architectures
• Evaluation of concepts and technologies
• Preparation of decision templates
• Architectural Decision Records (ADR)
• Coordination of implementation
• Review of implementations
• Documentation

• Supporting several agile teams in the new product development of an enterprise CRM and ERP application
• Web frontend development with Angular 17–21 (microfrontends, Domain-Driven Design, MDB-Bootstrap)
• Backend development with C# and .NET 7–10, MongoDB, MassTransit, RabbitMQ, Entity Framework Core, ElasticSearch (microservices, DDD)
• Technical feature design for assigned tasks
• Technical architecture design including forward-looking planning of AI-supported extension options
• Proof-of-concept implementation of AI-supported workflows with n8n and a locally hosted LLM (Ollama)
• Coordination with stakeholders and Product Owners, support with sprint planning

Result: Delivery of production-ready microservices and frontend modules in agile sprint cycles.

Environment: .NET 7–10 · Angular 17–21 · C# · TypeScript · OpenAPI · Microfrontends · DDD · Kubernetes · Docker · Helm · MS SQL Server · PostgreSQL · MongoDB · MassTransit · RabbitMQ · EF Core · ElasticSearch · Kibana · Azure DevOps · KeyCloak · n8n · Ollama

Seqnaut and Sycosm (WIP) – Embedded Audio DSP & Synthesis Platform

• Embedded audio DSP framework — synthesizers, effects, drum machines; modular audio graph
• Sequencing: PianoRoll, StepGrid, MasterClock; 8-voice polyphonic synth; CLI + OSC
• Transient Detector — dual-envelope adaptive threshold, guitar attack detection
• Hardware: Teensy 4.1, custom AFE design tapping BOSS GX-100; MIDI FX integration

Label: C++23, C, Python, CMake, FreeRTOS, Teensy 4.1 (ARM Cortex-M7), Teensy Audio (Stoffregen), DaisySP, I2S/SAI, OSC, MIDI

• Third-side integration development (REST: Amazon, OpenAI, Etsy, Shopware, Prestashop, etc)

• Performance optimization (Refactoring, SQL analysis)

• LMS development and deployment (Moodle)

• Cloud deployment automation (Amazon, Azure, Gitlab CI/CD, Docker)

• Backend and Frontend development (Laravel, VueJs)

• Collaboration with partner-developers (national and international)

• Shirtigo GmbH

• GRELLARTIG Digital GmbH

• velpTEC edutainment GmbH

• Introduction of giro card with Visa Debit card payment function (Combocard) for Commerzbank
• Support for the Card Processing Team (card processing; Clearing & Settlement incl. Dispute Management teams) in project work
• Creation of requirements for the Clearing & Settlement team, Authorization team (AZ) and Card Production team (KP) as part of a requirements catalog
• Requirements engineering (requirements management) with business contact person
• Participation in workstream meetings between VISA Inc., SRC Security Research & Consulting GmbH, Commerzbank AG and consulting firms (PwC and Senacor Technologies AG)
• Sprint planning with Jira tickets of type Task and use of Confluence
• Creation and presentation of weekly status reports for the Clearing & Settlement, Dispute Management and Reporting & GUI workstreams
• Coordination with network and file transfer teams for connection tests of new interfaces, e.g. setup of host keys, use of VPN or sipnet connections
• Coordination of the delivery of test files, e.g. exchange rates (FX rates; TC 56 of VISA rates), VISA Debit and Credit transactions (format Base II) and EANSS/ISS Base II clearing files to the Debit Clearing System (DCS)
• Conducting coordination meetings on the use of Visanet Settlement Service (VSS) and PSD2 reports and payment traffic statistics (ZVS)
• Organizing workshops with FiServ (First Data) on Dispute Management and complaint handling
• Setup and maintenance of Open Point List (OPL) in Confluence
• Technologies/methods: Jira, Confluence, giro card, VISA Debit card, payments, Kanban board, BIN, routing, CIQ, VPAY, ScalaTest, Dispute Management

• Visualization, search, and maintenance of various types of master data, providing them via a REST API, and implementing a variety of import and export formats.

• Implementation, architecture, project management.

• Technologies and tools: Angular (Signals, Signal Store, swagger-codegen, standalone components), Java, PostgreSQL, ActiveMQ, JavaScript, Spring Boot (Spring Core, Spring MVC, Spring Data/JPA via Hibernate, Spring Security, Spring Modulith), JUnit, Flyway, QueryDSL, GraphQL, REST API, Node, npm, Maven, SonarQube, GitLab CI/CD pipelines, Docker, Nexus, OpenAPI + Swagger UI, Keycloak, OAuth2, OpenID Connect, IntelliJ IDEA, Jira, Confluence.

• Consulting on processes and execution of load tests, performance tests and technical tests of various applications: web portal, file cloud, email, video conferencing, learning/course platform
• Creating the load test framework
• Programming test scripts (JMeter, Selenium)
• Developing batches to support the evaluation
• Executing load tests and evaluating results, including clear reporting, recommendations for action, and bottleneck analysis
• Managing tasks and collaboration to achieve meaningful load test results