"I live with data, for data — because everything is about data."
mindmap
root((Data))
Platform & Architecture
Medallion Architecture
ETL / ELT Pipelines
Data Warehouse Design
Data Science & ML
Statistical Analysis
Machine Learning
MLOps
DataOps & Governance
Data Quality
Data Catalog
Data Lineage
BI & Decision Systems
Power BI & DAX
KPI Definition
Reporting
Data Scientist with organization-wide ownership of analytics, data platform modernization, and enterprise data architecture in logistics and supply chain. Experienced in Medallion Architecture, scalable BI ecosystems, ERP integrations, and cross-functional team leadership. Focused on architecture-driven data operating models that improve reliability, traceability, and decision-making at scale.
A practical data operating model should not stop at reporting. It should move from source systems to trusted decision layers through a controlled pipeline:
Ingestion → Transformation → Quality → Catalog → BI
- Ingestion: ERP systems, operational databases, APIs, telemetry, and external sources are collected through Python-based services, orchestration pipelines, and integration workflows.
- Transformation: Raw and operational data is standardized, modeled, and prepared for analytics through SQL-first transformation patterns and dbt-based ELT workflows.
- Quality: Data quality rules, validations, freshness checks, and test scenarios are managed through Soda Core and dbt tests to improve trust in downstream datasets.
- Catalog: Metadata, lineage, ownership, and dataset documentation are centralized through OpenMetadata to make the platform easier to govern and scale.
- BI: Clean and controlled data is exposed to reporting and decision systems such as Power BI, Looker Studio, and operational dashboards.
This structure reflects how I approach sustainable analytics: not as isolated dashboards, but as a governed and scalable data platform.
project_name/
├── data/ # DVC-tracked, git-ignored
│ ├── raw/ # Immutable source data
│ ├── processed/ # Cleaned & transformed
│ └── features/ # Feature engineering outputs
├── notebooks/ # EDA & experimentation
├── src/ # Production source code
│ ├── data/
│ ├── features/
│ ├── models/
│ └── visualization/
├── tests/
├── configs/
├── monitoring/ # Grafana dashboards & alerts
├── deployment/ # Docker, cloud configs
└── docker-compose.yml
Template of choice: cookiecutter-data-science by @drivendataorg
