Karlsruhe, Baden-Württemberg, Germany
Hey there, I’m passionate about Open Source, Smart Home, Artificial Intelligence, and Python development. Whether I’m building intelligent automation platforms at work, developing retrieval-augmented generation (RAG) pipelines, or maintaining widely used open-source projects like Mealie-Bring-API (> 1.5 Million downloads), I thrive on creating impactful solutions. In my free time, I actively contribute to projects like Home Assistant, Mealie, and BorgBackup, as well as developing my own tools - from smart home integrations to intelligent energy controllers. I believe in the power of community-driven innovation and love exploring new technologies.
As a Site Reliability Engineer my team and I are responsible for designing, developing, maintaining and supporting our internal Middleware as a Service.
As a software developer for DE-CIX, I worked on automating the configuration of complex network topologies. DE-CIX uses large enterprise routers to distribute traffic around the globe. These routers need to be provisioned with their respective configurations. My team and I were working on implementing a software stack that listens for changes in a datacenter infrastructure management tool called NetBox and applies the changes to the routers. Along the way, it validates the data entered by the user, parses it, creates a diff between the configuration on the routers and the requested change using YANG models, and then applies it. While writing code for this mostly in Python, I came across a wide variety of tools to support my day-to-day work, such as Terraform, Kubernetes, Microsoft Azure, Apache Kafka, Gitlab CI/CD, containerization, Prometheus and Grafana to name a few.
My master’s thesis was developed in cooperation with DE-CIX and focused on the conception, implementation and evaluation of a model-driven network orchestrator for automated router configuration. The goal was to replace the previously used, resource-intensive and vendor-dependent orchestration platform with a lean, domain-specific solution tailored to the needs of a global Internet Exchange Point. The developed orchestrator derives deterministic device configurations from NetBox as the central source of truth, validates them against YANG models, performs a target-versus-actual comparison, and deploys changes to routers in a transaction-safe manner. As part of the thesis, I designed the architecture based on a clear separation of domain and infrastructure concerns and helped implementing the full orchestration workflow in Python. An exemplary feature implementation (Port Mirroring) demonstrated the extensibility and consistency of the chosen approach. The result is a production-ready orchestrator that increases transparency, reduces structural complexity and establishes technological sovereignty.
My bachelor thesis was developed in cooperation with DE-CIX and was concerned with automatically generating information-rich graphs from text-based information sources, keeping them up-to-date and making them available to employees of DE-CIX. The aim was to develop an abstract and interactive representation of the DE-CIX network, which depicts the routers, locations and regions involved as well as their connections to each other. This goal of the thesis at hand was achieved: a series of interactive and interconnected graphs are created, which can be accessed within the company and show the different components of the network and their relationships to each other. Furthermore, these graphs show additional details such as names, locations, coordinates, metrics, IP addresses and further links at the user’s request. The result of this work enables DE-CIX employees to obtain a graphical overview of their own network – in addition to the existing text-based sources – and thus expand their understanding of it. With this knowledge, network expansions can be better planned and disruptions can be resolved more quickly and efficiently. „The maps created make everyday network operations easier by illustrating the topology and showing dependencies. In addition to network operations, other teams will also benefit from these maps [...]“ (~Robert Finze, Head of the Network Operations Department at DE-CIX)
I started my career at AB+M when I was 17 years old, right after I finished high school and before I started studying computer science. As it was my first job, I learned a lot of the basics of software development and maintenance, such as working with GIT and SVN, SSH, Ansible, databases, CRM systems, Atlassian Jira & Confluence, and programming my first scripts in Bash, Python, JavaScript and Java, which were adopted by the whole company for everyday use in my colleagues. My main responsibilities at AB+M were: - 1st and 2nd level customer support - Administration and maintenance of local and customer VMs with Ansible - Designing, programming and supporting internal tools & software What the CEO Christopher Asp had to say about me (translated from German to English): "Thanks to his excellent comprehension skills, he was always able to grasp even difficult situations immediately and find excellent solutions quickly." "Mr. Schneider constantly showed a high level of initiative and identified with his tasks and our company in an absolutely excellent manner, whereby he also impressed us with his exemplary enthusiasm." "Mr. Schneider was particularly willing to learn. Even in situations with a heavy workload, he always proved to be highly resilient."
In this internship, I developed a web application for an internal tool that was used to collect data from customers on a given problem and store it in a central location for my staff to access and analyse using a tool that I had previously developed. The aim of this web application was to make this tool, which previously only ran on the command line, more user-friendly and more accessible. This goal was fully achieved: The result of this project was able to standardise, automate and speed up a process that was performed numerous times a day, from typically about ten to thirty minutes of repetitive manual work, to a quick query that finished in about 20 seconds and could be run in the background. It also visualised the results directly in the browser. This tool was used by the whole support team, speeding up workflows and helping colleagues who were not familiar with the command line enormously.