Ingest
APIs, relational DBs & files — extracted with resilient, checkpointed jobs.
Bangkok, Thailand · Open to opportunities
Siriwat
Wisetpakdeewong
> Data Engineer
I build and migrate large-scale data pipelines across cloud platforms — turning messy, 19-million-row raw data into reliable, automated flows with Airflow, Spark, Kafka and Databricks.
Turning raw data into reliable pipelines
I'm a Data Engineer with 2.5+ years of experience building and maintaining data pipelines across cloud platforms and large-scale migration projects.
I have a proven track record of delivering technical solutions that drive efficiency across complex data workflows — from architecting greenfield ETL pipelines to migrating thousands of legacy objects onto modern data platforms. I care about production-grade code, reproducible environments, and pipelines that run unattended.
Most recently I've been working on enterprise data-platform migrations for one of Thailand's largest banking technology groups — converting legacy shell & IBM DataStage logic into Databricks, building automation that removed hours of manual work, and shipping ahead of schedule.
- Cloud data migration across GCP, Azure & AWS
- Batch & real-time pipelines with Airflow, Spark & Kafka
- Data modeling — SCD Type 2, Medallion, lakehouse
- Legacy modernization — shell & DataStage → Databricks
// education
B.Eng. Aircraft Electronics Engineering — 1st Class Honours (GPA 3.52)
Civil Aviation Training Center (CATC) · 2020 — 2024
High School
Howick College, New Zealand · 2018 — 2020
// languages
Thai (Native) · English (Fluent) — TOEIC 930
A pipeline, end to end
How I take data from raw source to business value — watch the DAG run, and click any tool tag to see it used in a real project.
Process
Clean, normalize & transform at scale — config-driven, production-grade code.
Orchestrate
Scheduled, backfill-ready DAGs with error handling and zero manual steps.
Store
Landing zones, lakehouses & warehouses — Bronze / Silver / Gold layers.
Serve
Validated, reconciled data ready for analytics and business consumption.
Tools I build with
The technologies I reach for across ingestion, processing, storage and orchestration — click any chip to see the projects that use it.
</>
Languages
- Python
- SQL
- Bash / Shell
- PySpark
⚡
Processing & Big Data
- Apache Spark
- Pandas
- Delta Lake
- dbt
- psycopg2
⟳
Orchestration & Streaming
- Apache Airflow
- Azure Data Factory
- Apache Kafka
- Confluent
- ZooKeeper
☁
Cloud Platforms
- GCP · BigQuery
- GCP · Cloud Storage
- Azure · Databricks
- Azure · ADLS Gen2
- Azure · SQL
- Microsoft Fabric
- AWS · S3
- AWS · Glue
- AWS · Athena
- AWS · Redshift
◱
Databases
- PostgreSQL
- MongoDB
- MySQL
- Cassandra
- SQL Server
◈
Modeling & DevOps
- SCD Type 2
- Medallion Architecture
- Data Lakehouse
- Docker
- Git
- IBM DataStage
- Tableau
Work experience
Data Engineer
Inteltion
Enterprise data-platform migrations for Kasikorn Business Technology Group (KBTG).
Modern Data Platform Migration
- Led a sub-team of 6 within a 20+ person project, delivering ~1,000 of 5,000 objects ahead of schedule.
- Built an automated sample-data extraction script that removed manual work and fixed mixed load-frequency issues, saving ~2–3 hrs/day across ~70% of the project.
- Established communication standards & execution-order protocols across all 5 sub-teams to keep work synchronized.
- Trained 5 new joiners to independent productivity within 2 weeks; resolved ~50 PROD incidents through final sign-off.
Data Logic Migration
- Built a flagging script that auto-identified 450 of 900 objects (50%) needing no manual intervention, focusing the team's effort.
- Conducted job-chain analysis across shell scripts & IBM DataStage, mapping nested dependencies to set correct execution order across 900 objects.
- Converted population logic from shell/DataStage into Databricks notebooks with full functional equivalence to the legacy system.
- Databricks
- IBM DataStage
- Shell
- SQL
- Leadership
Data Engineer Internship
Bluebik Digital
Big Data Migration Across Cloud Platforms
- Modeled a 19M-row raw dataset into normalized tables using SCD Type 2 and mapping tables to resolve conflicts and preserve history.
- Built an automated daily batch pipeline with Airflow & PySpark processing ~50K rows/day into a GCS landing zone with backfill recovery, orchestrated into BigQuery.
- Refactored procedural scripts into dynamic, config-driven OOP classes, cutting pipeline runtime by 30%.
- Implemented a scheduled Azure Data Factory pipeline into ADLS Gen2 with Databricks reconciliation that caught non-ASCII encoding mismatches before loading Azure SQL.
- Containerized Airflow & Spark via Docker for full pipeline testing at zero cloud cost.
- Airflow
- PySpark
- GCP
- BigQuery
- Azure Data Factory
- Databricks
- Docker
Data Engineer Internship
NECTEC — National Electronics & Computer Technology Center
Data Pipeline Development
- Independently architected, built & delivered a greenfield ETL pipeline for medical product data — bringing a previously unhandled flow into fully automated daily operation.
- Extracted & denormalized ~450K rows across 4 PostgreSQL tables via psycopg2 & Pandas into a clean, consolidated dataset.
- Applied extensive Pandas transformation & cleaning to ensure high data consistency and accuracy.
- Orchestrated daily transformation workflows with Airflow, writing the latest output to CSV for easy downstream access.
- Airflow
- Python
- Pandas
- PostgreSQL
Freelance projects
Data Engineer Freelance
NECTEC — National Electronics & Computer Technology Center
A post-internship engagement continuing the NECTEC medical-product data pipeline (MedQ), rebuilt to client-level coding standards.
Database Migration — PostgreSQL → MongoDB
- Extended the internship pipeline into a production-grade PostgreSQL → MongoDB migration, extracting & denormalizing multiple relational tables via psycopg2 into a MongoDB-optimized document schema.
- Added selective row updates, checkpoint tracking and enhanced error handling for resilient, unattended daily runs.
- Orchestrated daily automated runs with Airflow; loaded to MongoDB per client-specified pipeline structure & data format.
- Implemented PostgreSQL ↔ MongoDB reconciliation to verify accuracy and consistency end-to-end.
- Airflow
- Python
- Pandas
- PostgreSQL
- MongoDB
Personal projects
Hands-on data engineering builds — every card opens a full case study with architecture diagrams, the story behind the build, and live GitHub stars. Pulled from my GitHub.
Medallion Data Lakehouse
End-to-end Azure lakehouse implementing the Bronze / Silver / Gold Medallion architecture. Data Factory dynamically extracts source tables into Delta Lake on ADLS Gen2; Databricks and dbt transform through curated layers, with secrets in Key Vault.
- Azure
- Data Factory
- Databricks
- dbt
- Delta Lake
Multi-Cloud Data Migration Pipeline
Multi-cloud big-data migration: normalizes the liquor-sales dataset into 11 tables and loads to GCS + BigQuery via Airflow & PySpark, then onward to ADLS Gen2 + Azure SQL with Data Factory orchestration and Databricks reconciliation. The real-world build behind my Bluebik internship.
- PySpark
- Airflow
- GCP
- Azure
- Databricks
- Docker
Realtime Data Streaming
End-to-end streaming pipeline: an Airflow DAG pushes API data into Kafka; Spark Structured Streaming processes it and writes to Cassandra — with Confluent Control Center & Schema Registry for monitoring. Fully containerized.
- Kafka
- Spark
- Cassandra
- Airflow
- Confluent
IoT Event Streaming Pipeline
Simulates real-time IoT telemetry (weather, traffic, GPS, vehicle, emergency) streamed through Kafka → Spark → S3 as Parquet. AWS Glue catalogs & transforms the data, queried in Athena and loaded into Redshift for analytics.
- Kafka
- Spark
- AWS S3
- Glue
- Athena
- Redshift
Postgres → MongoDB Migration
Airflow-orchestrated ETL that extracts from a PostgreSQL source, transforms with Pandas, and loads into a MongoDB document model — with end-to-end source ↔ target reconciliation. The pattern behind my NECTEC medical-data pipeline.
- Airflow
- PostgreSQL
- MongoDB
- Pandas
- Docker
NASA Events ETL
Pulls data from NASA's public API and fans it out to multiple destinations — PostgreSQL, MySQL, MongoDB and AWS S3 — through a single configurable, Airflow-orchestrated ETL.
- Airflow
- AWS S3
- PostgreSQL
- MySQL
- MongoDB
ExcelToPostgres — Code Quality
A refactoring study: transforms a procedural Excel → PostgreSQL loader into clean, config-driven object-oriented code with a production-grade project structure and best practices.
- Python
- OOP
- Refactoring
- psycopg2
Wikipedia Web Scraping
A focused mini-project scraping tabular data from Wikipedia with BeautifulSoup and shaping it into a clean, analysis-ready pandas DataFrame and CSV export.
- Python
- BeautifulSoup
- Pandas
(0 rows) — no projects match this filter.
Certifications
Click a certification to see what it covers.
Databricks Certified Data Engineer Associate
Databricks
verify credential ↗Microsoft Certified: Fabric Data Engineer Associate
Microsoft · DP-700
verify credential ↗Microsoft Certified: Security Operations Analyst Associate
Microsoft · SC-200
verify credential ↗SQL (Advanced)
HackerRank
verify credential ↗
08 / contact
Let's build something
Let's build something
data-driven together.
I'm open to Data Engineering roles and collaborations. The fastest way to reach me is email — I'll get back to you.