CV

Eddie Beloiu

Eddie Beloiu

Contract Data Engineer. I migrate legacy and SAP systems into scalable, AI-ready data platforms. Based in Munich, Germany. English and German.

Experience

Jan 2026 - Jun 2026
Remote

Data Engineer: SAP S/4HANA to PLM Migration

Global food manufacturer (contract)

International SAP S/4HANA to SAP PLM master-data migration for a global food manufacturer, engineered end to end and delivered to production.

  • Built and shipped the Python ETL that migrated raw-material and finished-product master data into the production SAP PLM system
  • Implemented bill-of-material to recipe logic, including the recipe water-content calculation
  • Safeguarded the production cut-over with batched SQL loads, reconciliation audits, and automated tests
  • Coordinated validation and sign-off in German and English across teams in the Netherlands, Germany, and Spain
PythonpandasopenpyxlSQL ServerSAP S/4HANASAP PLMETLGit
May 2025 - present
Munich

Contract Data Engineer

Self-employed

Independent contract data engineering for German Mittelstand clients: migrating legacy and SAP systems onto governed, cloud-native platforms with AI-ready analytics layers.

  • Cloud platform consulting and MLOps infrastructure for AI applications
  • Agentic workflows with LangChain/LangGraph
  • ETL pipeline development and Python automation
  • AI model evaluation and integration with Pydantic and MLflow
  • Lakehouse architecture design and data quality rule implementation for client data products on Azure Databricks
  • Azure DevOps CI/CD pipelines and Terraform IaC for repeatable environment provisioning
  • Active member of GDG Munich (Google Developer Group)
PythonLangChainLangGraphRAGAzure DatabricksPySparkdbtMLflowPydanticSAP S/4HANASAP PLMAzure DevOpsJiraDockerGit
2024 - 2025
Nuremberg

Data Engineer

Trevisto AG

End-to-end data platform engineering at a mid-sized technology company.

  • Architected end-to-end data platform (Snowflake, Python)
  • Data Vault 2.0 modelling for the analytics layer
  • Designed data governance processes and implemented data quality rules across ingestion and integration pipelines
  • Built Fabric Retrieval System with REST API and change tracking/alerting
  • Microsoft Fabric Data Warehouse prototype for finance and controlling
  • Integrated Google Cloud and BigQuery into existing pipelines to pull marketing data for ML models and analysis/dashboards
  • Agile delivery via SCRUM with JIRA for backlog management, sprint planning, and release tracking
SnowflakePythonAzure DatabricksPySparkdbtData Vault 2.0Microsoft FabricGoogle CloudBigQueryTerraformAzureREST APIJiraGitSASOracle SQL
2022 - 2024
Munich

Data Engineer

Bertelsmann

Data platform and BI engineering for two Bertelsmann group companies, working across marketing analytics, lakehouse architecture, and ML deployment.

  • Designed Delta Lake lakehouse platform (Azure Databricks) for 23M+ records, including medallion architecture and data integration interfaces for downstream services
  • Implemented PySpark-based data quality rules and validation checks across Bronze/Silver/Gold layers
  • Deployed Bayesian ML model (PyMC3) on Azure ML Studio with full CI/CD and inference pipeline; 8% deviation from actual ROAS in A/B tests
  • Built FastAPI REST API for downstream service integration
  • Developed Power BI semantic models and DAX measures for marketing performance reporting and campaign analytics
  • Integrated Google Cloud and BigQuery into existing pipelines to pull marketing data for ML models and analysis/dashboards
  • Automated Tableau backup and version control (Python, Git), adopted as standard across multiple departments
  • Developed GenAI NLP tools: sentiment analysis (Hugging Face Transformers) and text summarisation (BART) with batch processing and quantisation
  • Ran monthly Python training sessions for 8 team members
Azure DatabricksDelta LakeApache AirflowPySparkPythonSQLdbtFastAPIPyMC3Azure ML StudioPower BITableauGoogle CloudBigQueryJiraSASOracle SQL
2022 - 2022
Munich

Sales Data Analyst

Tradelink GmbH

Fixed-term internship at a remote-first logistics-tech startup.

  • Python-based tools for data analysis and automation
  • Lead generation via automated and manual web scraping
  • CRM (HubSpot) workflow and tracking dashboard setup
PythonHubSpotJiraMiro
2012 - 2021
USA

Technical Lead

Clear Earth

Technical Lead at a small company, overseeing full stack development across web applications, databases, and DevOps infrastructure while managing business operations.

  • Full stack development: Java, Angular, React, Spring Boot, Node.js
  • Databases: PostgreSQL, MySQL, REST APIs with TypeScript
  • DevOps: Docker, Kubernetes, CI/CD pipelines, Git
  • Integrated Google Cloud and BigQuery into existing pipelines to pull marketing data for ML models and analysis/dashboards
  • Business leadership: company formation, client acquisition, financial management, and team coordination
JavaSpring BootAngularReactNode.jsTypeScriptPostgreSQLMySQLGoogle CloudBigQueryDockerKubernetes
2007 - 2012
Various

English Language Teacher

Self-employed

Taught English as a foreign language across four countries, developing cross-cultural communication skills and adapting to diverse learning environments.

  • Delivered language instruction to students of varying ages and proficiency levels
  • Developed customized lesson plans and educational materials
  • Adapted teaching methodologies to different cultural contexts and educational systems

Skills

Data Platforms
Azure DatabricksSnowflakeDelta LakeMicrosoft FabricAzure Cloud
Data Engineering
Apache AirflowAzure ADFdbtdltPySparkETL/ELTData Vault 2.0
AI & ML
RAGLangChainLangGraphLlamaIndexMLflowPyMC3Hugging FaceScikit-learn
Programming
PythonSQLTypeScriptJavaFastAPINode.jsSASOracle SQL
DevOps & Tools
DockerKubernetesGitCI/CDAzure DevOpsJiraTerraform
Compliance & Governance
GDPRData GovernanceAudit TrailsData Quality

Education

Bachelor of Science in Computer Science (incomplete)

Oregon State University · 2020 - incomplete

Bachelor of Science in Natural Sciences

Texas A&M University · 2008 - 2012

Languages

EnglishNative
GermanFluent

Certifications

dlt ELT Advanced Specialist

dltHubAug 2025

dlt ELT Specialist

dltHubAug 2025

Master of ELT Best Practices

dltHubJun 2025

Data Vault 2.0

Data Vault InstituteNov 2024

Hands-On Essentials: Data Warehousing Workshop

SnowflakeSep 2024

dbt Fundamentals

dbt LabsAug 2024

Databricks Lakehouse Fundamentals

DatabricksJan 2022

Linux Administration

CompTIAJan 2022

Python Certification

Python InstituteAug 2019

Open Source

Local SLM Data Cleaner

Fine-tunes a 0.6B Qwen on-device (LoRA + MLX) to normalize messy SAP master data. Runs locally, no cloud API, no GPU rental, no GDPR exposure. Where rule-based pipelines need a new rule for every spelling or alias, the model absorbs the long tail in one pass.

PythonMLXQwen3llama.cpp

Enterprise Context Layer

Captures scattered, contradictory company knowledge (Slack, Jira, Confluence, etc) that search can't touch. LLM agents synthesize cited knowledge into a Markdown git repo; parallel workers coordinate via git's push-rejection as a built-in sync. No Redis, no message broker needed. Loads narrow skill files on task match instead of one bloated system prompt. Runs on Claude Code, Codex, OpenCode, Ollama, or Pi.

PythonAgent SDKGitYAML

Let's talk about your platform.

Munich-based, working across the DACH region. Every engagement includes documentation and handover.