Data platforms
Project Siphon
Real-time change data capture for every GitLab deployment.
Production · GitLab
- PostgreSQL
- NATS JetStream
- Go
- Apache Iceberg
100%
of database changes captured, closing a 30–40% gap
DATA SUPERHERO
> hello, I'm
Ved Prakash.
I build data and AI platforms that last, and lead the teams that run them.
Principal Data Engineer at GitLab with 15+ years building enterprise data and AI platforms at petabyte scale. I lead Project Siphon, GitLab's real-time CDC platform, and own AI governance and FinOps. Based in Almere, Netherlands.
- Experience
- 15+ years
- Role
- Principal Data Engineer · GitLab
- Recognition
- Data Superhero Council · 1 of 17
- Based in
- Almere, NL
About me
Data platforms, governance and the people who run them
Ved Prakash is a Principal Data Engineer at GitLab with 15+ years building enterprise data and AI platforms at petabyte scale. He leads Project Siphon, GitLab's real-time change data capture platform built on PostgreSQL, NATS JetStream, Apache Iceberg and Snowflake, and designed GitLab's company-wide governance framework for Snowflake Cortex AI. A multi-year Snowflake Data Superhero and one of 17 members of the global Data Superhero Council, he speaks at Snowflake Summit, Berlin Buzzwords and DSC Europe, and writes The DataAI Chronicles. He lives in Almere, Netherlands.
Full story →2024 – Present/Remote, Netherlands
Principal Data Engineer
GitLab · Enterprise Data Platform · Project Siphon DRI
2022 – 2024/Remote, Netherlands
Staff Data Engineer
GitLab · Data Platform Team
2021 – 2022/Remote, Netherlands
Senior Data Engineer
GitLab
Impact
Measured outcomes
- total annual value from Project Siphon
- ~$1M/yr
- in potential AI overruns prevented
- $100K+/mo
- database change capture, closing a 30–40% gap
- 100%
- pipeline runtime reduction, 1,456 compute hours a year saved
- 62%
- migrated from on-premise Kognitio to Snowflake on AWS
- 100TB+
Recognition & certifications
Recognised by the community
- Snowflake Data SuperheroMulti-year
- Data Superhero Council1 of 17 on the global council
- Snowflake Certified Architect
- Snowflake Certified Developer
- Professional Data Engineer
- Author, The DataAI Chronicles
- Conference speaker
Selected work
Platforms, governance and products
Governance & AI
Cortex AI Governance & FinOps
Governance that lets a company adopt AI without runaway spend.
Production · GitLab company standard
- Snowflake Cortex
- RBAC
- Credit monitoring
- Alerting
$100K+/mo
potential AI overruns prevented
Products
PostCraft
A LinkedIn post editor whose formatting survives the paste.
Live · free, no login
- React
- TypeScript
- Unicode text styling
- AI rewrites
Speaking
On stage
June 2026AI256
Snowflake Summit 2026 · San Francisco
How GitLab Saved 90% in Engineering Costs by Keeping AI Inside Snowflake
With Radovan Baćović, GitLab
Processing 75,000+ Zendesk tickets for ~$1,500 with Snowflake Cortex AI versus a $73,000+/year external pipeline.
Also delivered
June 2026 · Snowflake Summit 2026
Beyond Passwords: Strong Authentication for Snowflake Workloads
15 Oct 2025 · DSC DACH
Talk to Your Data: Implementing Natural Language Analytics with Snowflake Cortex
17 Jun 2025 · Berlin Buzzwords
Siphon: Modern Data Stack with SF-CH & Iceberg
Writing
The DataAI Chronicles
The Quietest Process in Airflow: What the Triggerer Did for 200 Parallel Dataproc Jobs (opens in a new tab)
big datagoogle cloud platformkubernetes
7 min
Connecting Atlan to Snowflake’s Ontology: Tags, Lineage, and Semantic Views Across the Boundary (opens in a new tab)
data engineeringontologydata governance
10 min
The Semantic Layer: Semantic Views, dbt Metrics, and Business Definitions (opens in a new tab)
data engineeringdata governancedbt
7 min
Beyond work
Life across borders
Lived in the Netherlands, the United Kingdom and India; spoken in five more countries.
More beyond work →