I design, build and operate data pipelines that move your ERP and business data reliably into your data warehouse — orchestrated with Apache Airflow, versioned like software, kept as simple as possible.
Technologies I work with every day
Data platforms and the infrastructure beneath them — from one source, as code, with knowledge transfer.
From use case to pipeline: I move data from SAP and non-SAP systems into your data warehouse — reliably, tested and documented. And I only build what demonstrably delivers value.
The foundation for over ten years: Infrastructure as Code, CI/CD pipelines, Kubernetes and cloud architectures on Azure and AWS — automated, not hand-clicked.
Your business sees dashboards, reports and use cases. For those to work reliably, solid craftsmanship is needed underneath: data models, pipelines, quality assurance and operations. That heavy lifting is exactly my job.
No over-consulting, no over-engineering: I start from your requirements and use cases — AI-assisted analysis on request — and only build what demonstrably delivers value. For the Mittelstand that means simple, proven technology your team can operate itself.
Six building blocks that work individually or as a complete package.
Before a single line of code: workshops, stakeholder interviews and AI-assisted requirements analysis. We prioritize use cases by business value — and cut what doesn't deliver any. The result is a backlog that pays off.
Extraction from SAP ERP, S/4HANA and BW — via ODP, CDS Views or SLT — into your target warehouse. Incremental, delta-enabled and robust against structural changes in the source system.
Architecture and build-out of your data warehouse — with SAP BW/4HANA, SAP Datasphere or a lean cloud warehouse. As simple as possible: layered models and data modeling your team understands and can evolve on its own.
Orchestration as code: DAG design, sensors, backfills, SLAs and alerting. Airflow deployments including CI/CD, secrets management and monitoring — sized to your needs, not to buzzwords. And because everything is code, modern AI tooling plays to its full strength here — reading, extending and testing pipelines directly. An edge point-and-click tools can't match: code-first is basically unbeatable.
Out of click-pipelines, into orchestration as code: I analyze your ADF pipelines, migrate them to Apache Airflow in a structured way, and make them testable, versionable and cheaper to run.
Trust in numbers comes from control: automated data quality checks right inside the pipelines, clear ownership, data catalog and lineage — pragmatically sized instead of governance bureaucracy.
Pipelines are software — they get version control, tests, CI/CD and monitoring. And technology is a means to an end: I recommend the simplest solution that reliably meets your requirements, and exactly as much consulting as the project really needs. Especially in the Mittelstand, what wins is what your own team can understand and operate.
Over ten years of automation experience — the same discipline that powers my data platforms today.
Terraform modules that make your Azure and AWS environments reproducible — reviewable, testable, documented.
GitHub Actions and Azure DevOps: from commit to deployment without manual steps, with quality gates.
AKS clusters, Helm charts and container strategies — including running Airflow and data services on Kubernetes.
Architecture reviews, cost optimization, security hardening and compliance for your cloud landscape.
Free intro call, a close look at systems and goals, an honest assessment — even if the answer is "you don't need this".
Iterative delivery as code: short feedback cycles, transparent communication, working increments instead of slide decks.
Documentation, pair programming and workshops until your team can evolve the platform on its own.
My name is Fethullah Misir. As a Data & DevOps Engineer, I help companies modernize their data and cloud platforms — from SAP extraction and Airflow orchestration down to the infrastructure beneath. My focus: pragmatic, sustainable solutions and knowledge transfer into your team.
A testing framework for isolated database tests in Spring Boot applications. Uses PostgreSQL template databases for fast test isolation (~50ms per test).
View on GitHubAnswers to the questions that come up most in intro calls.
Usually via ODP (Operational Data Provisioning), CDS view extraction or SLT replication. Which route fits depends on your source system, delta requirements and licensing — exactly what we clarify in the intro call.
Airflow is the de-facto standard for data orchestration: pipelines as Python code, versioned and testable, with retry logic, backfills and a huge ecosystem of integrations — without vendor lock-in.
That depends on your system landscape, team and budget. In SAP-heavy landscapes often SAP BW/4HANA or Datasphere, otherwise a lean cloud warehouse. I recommend the simplest solution that reliably meets your requirements — the Mittelstand rarely needs big-tech stacks.
Often yes: Airflow makes orchestration versionable, testable and transparent — and frequently lowers operating costs. I analyze your ADF pipelines and give an honest recommendation. Sometimes it is: stay on ADF.
Project-based and flexible — as a temporary part of your team or for clearly scoped initiatives with defined deliverables. I'm based in Schwäbisch Gmünd near Stuttgart and work remotely as well as on-site, across Germany, Austria and Switzerland.
Yes, knowledge transfer is a fixed part of my work. Workshops and pair programming on Airflow, dbt, Terraform and Kubernetes, so your team can continue independently.
Free intro call — 30 minutes, honest assessment, no sales pressure.