Freelance data platform engineer · Munich
I turn legacy systems into modern, AI-ready platforms.
I'm Eddie. I build the pipelines, orchestration, and governance that let AI agents operate as a unified workforce overseen by human experts. More than ten years of software development experience, five years teaching in four countries and three years of production data engineering.
Two problems, deeply.
I focus on what I do best, not everything. These are the two specialties clients hire me for again and again. They aren't mutually exclusive, as they stem from the same engineering principles.
Legacy-to-modern transformation
Moving companies off SAS, SQL Server, SSIS, SAP, and on-prem warehouses onto governed, cloud-native lakehouses. All while keeping the business running without ending up with something only the freelancer can maintain.
- Legacy assessment & migration roadmap
- Lakehouse architecture: Databricks, Delta Lake, Snowflake
- ETL/ELT with dlt, dbt, Airflow
- Zero-downtime cutover & validation
- Documented handover from day one
Agentic integration with existing systems
Putting AI agents to work inside the tools you already have. Persistent assistants, custom Skills, and MCP connectors that reduce manual ops instead of adding another dashboard nobody opens.
- Persistent agents with compounding memory
- Custom Skills & MCP connectors to your stack
- Scheduled routines that run unattended
- Agents paired with (not replacing) your team
- Audit-trail built in
Cloud data platforms shipped across 4 industries.
Infrastructure cost reduction on a warehouse re-platform.
Years building and shipping software, three of them in data.
Currently mid-flight on an international S/4HANA migration.
Proof, not adjectives.
While client names have been withheld for privacy, the data and methodologies are real. The write-ups are archived under Notes.
downtime
Re-platforming a legacy ERP data estate to the cloud
Database design, Python development, and bilingual coordination across an international S/4HANA programme. Built data pipelines to handle schema mapping, profiling, and validation so the business never stops while the platform moves underneath it.
Retiring legacy SAS analytics for a governed lakehouse
Moved scattered, hard-to-maintain SAS workflows onto a modern warehouse with automated quality checks and a paved path the in-house team owns. Cut infrastructure costs and 60% faster reporting cycles.
A low-risk on-ramp to bigger work.
Start small and scoped. Every step credits toward the next, and every engagement ends with documentation and handover.
Quick scan
One focused day. I review one specific area e.g. a pipeline, a schema design, or a single integration. Then I deliver a written summary of what I found, what needs attention, and what can wait.
Production-readiness audit
One week. I review your data platform, RAG system, or ML pipeline and deliver an assessment with the five highest-impact issues and a prioritized fix plan. Includes a one-hour walkthrough with your team.
Working POC
Up to two weeks. I build a working, documented proof of concept for your data platform problem: pipeline, transformation, lakehouse setup, or AI integration. You leave with code, a short write-up, and a clear path to production.
Build & migrate, then care
Full transformation or agentic integration on a flexible day rate, then an optional retainer to keep it healthy.
Eddie is a highly skilled data engineer with deep expertise in modern data platforms. He successfully led the migration of our legacy data warehouse to a modern cloud-based architecture, implementing best practices in data modeling and pipeline orchestration. His ability to balance technical excellence with business requirements, combined with his strong communication skills, made him an invaluable asset to our team. I highly recommend Eddie for any complex data engineering projects.
Working with Eddie on our data lakehouse implementation was exceptional. His expertise in Delta Lake, Databricks, and Airflow enabled us to build a scalable, production-ready platform that handles millions of records efficiently. Eddie's systematic approach to data governance and quality assurance, combined with his ability to mentor junior team members, significantly improved our data engineering capabilities. He delivers high-quality work on time and is a pleasure to collaborate with.
Supporting an international SAP S/4HANA migration through database design, Python development, and bilingual coordination for a global consumer goods manufacturer. On the side: persistent AI agents for individuals and small teams.
Eddie Beloiu
I've been building data platforms for the last three years, mostly for German companies. Before that, ten years of full-stack engineering. The data work is more recent, the engineering instincts are not.
And before any of that, I taught for five years in four countries on two continents. That's where I learned the hardest part of any complex thing is making it understandable. I write the doc, I do the handoff, which is exactly what makes AI legible for the technical and non-technical people in the same room.
Have a legacy system to move,
or agents to wire in?
I'd love to hear from you if you need help with a data platform project, want to set up AI-augmented workflows for your team, or need hands-on guidance on how to use these tools in your day-to-day work.