Data Engineer | Building end-to-end ELT pipelines | Python • SQL • PySpark • Databricks • dbt • Airflow • Docker

🎯 I build end-to-end ELT pipelines, design analytics data warehouses, and clean and transform data at scale with PySpark on Databricks. I bring a strong analytics background (SQL, Power BI) and keep deepening my data engineering skills through hands-on projects — currently focused on PySpark, Delta Lake, and CI/CD.

I'm a Data Engineer with a strong analytics background, building real-world data infrastructure projects. My focus is on end-to-end ELT pipelines, dbt transformations, and cloud data warehouses (BigQuery, PostgreSQL). I believe in writing clean, tested, and documented code—and I practice this in every project.

Current Location: Utrecht, Netherlands
Learning now: PySpark & Databricks (hands-on Bronze→Silver cleaning projects) → CI/CD → next big build: an end-to-end lakehouse pipeline (Databricks + Delta Lake + dbt + Airflow + AWS)

• Languages: Python 3.11, SQL (Spark SQL, BigQuery, PostgreSQL)
• Processing: PySpark (DataFrame API), Databricks, Delta Lake (medallion architecture)
• Transformation: dbt (dbt-BigQuery), SQL for analytics
• Orchestration: Apache Airflow 2.9+
• Cloud Warehousing: Google BigQuery, PostgreSQL
• Storage: Parquet, Delta, Google Cloud Storage (GCS)
• Containerization: Docker, Docker Compose

• Parquet, CSV, JSON
• REST APIs (Python Requests)
• Open-Meteo API (weather data enrichment)

• Git & GitHub (version control, CI/CD)
• Linux / Bash scripting
• Jupyter Notebooks (EDA & prototyping)

Technologies: Airflow 2.9 | dbt | BigQuery | PostgreSQL | Python | Docker

A complete ELT analytics pipeline for U.S. domestic flight delays using real data from BTS (Bureau of Transportation Statistics).

What I Built:

• ✅ Python ingestion scripts (Parquet generation, API enrichment, data cleaning)
• ✅ PostgreSQL staging (local data landing zone)
• ✅ BigQuery transformation (dbt: 5 staging + 5 mart models)
• ✅ Apache Airflow orchestration (daily @ 9 AM UTC)
• ✅ Data quality layer (15+ dbt tests: unique, not_null, accepted_range, composite keys)
• ✅ BI dashboard with 4 analytical views

Key Metrics:

• Dataset size: 2-3 GB/month, 20M+ flight records analyzed
• Pipeline runtime: ~15 minutes
• Query optimization: Leveraged BigQuery partitioning & clustering

Key Findings:

• Identified airport-level inefficiencies as primary delay driver (stronger than weather/holidays)
• Applied BigQuery partitioning & clustering to reduce query scan costs
• Weather moderately affects delays; holiday periods show stable scheduling

👉 View Repository

Technologies: BigQuery | GCS | SQL

Hands-on project comparing 3 table strategies for NYC Yellow Taxi data (20M+ trips, Jan–Jun 2024).

What I Learned:

• External tables vs. regular tables (storage vs. query trade-offs)
• Columnar storage impact on query costs
• Partitioning effectiveness: 12x cost reduction on filtered queries
• When to use clustering with partitions

Results:

• Same query on non-partitioned table: 310 MB scanned
• Same query on partitioned table: 26 MB scanned
• Practical demonstration of BigQuery optimization

👉 View Repository

Technologies: PySpark | Databricks | Delta Lake | Spark SQL

Six hands-on Bronze → Silver cleaning pipelines on Databricks, following the medallion architecture. Practice projects — each one a full raw-to-clean pass on a different messy dataset (e-commerce orders, cafe sales, Netflix titles, gym sessions, raw event logs, IoT device events).

Techniques covered:

• ✅ Schema-on-read (StructType / DDL), clean-then-cast, try_* + coalesce for mixed formats
• ✅ Parsing raw unstructured logs with regexp_extract (chose regex over an LLM for determinism)
• ✅ Timezone normalization (CET → UTC), complex types (struct / array / map, withField)
• ✅ Grain-aware deduplication, null vs. invalid-value policies, idempotent Delta writes

👉 View Repository

• ELT/ETL Pipelines: From raw data to production-ready analytics
• dbt Best Practices: Staging layers, mart models, comprehensive testing
• SQL Performance: Query optimization, BigQuery partitioning/clustering
• Data Quality: dbt tests, schema validation, anomaly detection
• Orchestration: Airflow DAG design, error handling, monitoring
• Docker: Local reproducibility, image optimization
• Code Quality: Clean SQL/Python, documentation, version control

• ✅ DataTalks.Club Data Engineering Zoomcamp — Docker, dbt, BigQuery, Airflow fundamentals
• 🔄 PySpark & Databricks — Spark programming course + 6 published hands-on cleaning projects
• ⏭️ CI/CD for data pipelines — GitHub Actions, pre-commit, detect-secrets
• ⏭️ Flagship project: end-to-end lakehouse pipeline (Databricks + Delta Lake + dbt + Airflow + AWS S3)
• 🎯 Target certification: Databricks Certified Data Engineer Associate

Best for: Data engineering interviews, portfolio reviews, hiring managers

📌 Pinned Repositories:

1. end-to-end-flight-delay-pipeline — End-to-end ELT pipeline (Airflow + dbt + BigQuery)
2. databricks-pyspark-data-cleaning — PySpark Bronze→Silver cleaning on Databricks (6 projects)
3. bigquery-taxi-data-warehouse — Data warehouse design & optimization
4. gz-dbt-repository — dbt analytics pipeline (staging + marts + tests)
5. sql-financial-analytics-pipeline — Advanced SQL transformations
6. sql-order-analytics — Window functions & analytical SQL

✅ Real-world projects — Not toy datasets. Real data, real pipelines, real trade-offs.
✅ Data quality-first — Every pipeline includes comprehensive testing.
✅ Clear documentation — READMEs that explain the "why," not just the "what."
✅ Optimization mindset — BigQuery cost reduction, SQL efficiency, Airflow reliability.
✅ Learning in public — Active in DataTalks.Club community, documenting learnings.

💼 LinkedIn: kenan-tufan-k-263000308
📬 Email: kenantkurt@gmail.com
🐙 GitHub: @Kenantkurt
📍 Location: Utrecht, Netherlands

• 💼 Data Engineering roles: Junior to mid-level
• 🤝 Collaborations: Open-source data projects, learning groups
• 💬 Discussions: Data pipelines, dbt best practices, SQL optimization

Let's talk data! Feel free to reach out on LinkedIn or email. I'm always eager to discuss data infrastructure, ask questions, and collaborate on interesting problems.

Last Updated: July 6, 2026
Status: Actively learning & building 🚀