Principal Data Architect

Syndesus•San Jose, CA

6h•Hybrid

About The Position

We are seeking a highly experienced, hands-on Principal Data Architect to define the strategic vision for a next-generation, petabyte-scale data platform and lead the transition from centralized to decentralized data management. This is a principal-level role — the bar goes well beyond senior. We are looking for someone with demonstrated long tenure at complex, large-scale data environments who has driven real architectural transformation, not just advised on it. The core mandate is data decentralization — moving from a monolithic data warehouse model to a Data Mesh paradigm — alongside modernizing data security, governance, and real-time capabilities. Consulting backgrounds and serial short-tenure candidates will face significant additional scrutiny; we want people who have gone deep and have the organizational impact to show for it.

Requirements

10+ years of software and data engineering experience, with a significant portion in technical leadership or architecture
FAANG or Big Tech background strongly preferred
Proven track record at large-scale enterprise environments — petabyte-scale data infrastructure is the baseline expectation
Long tenure at key roles is a strong positive signal; we are not looking for contract-to-contract backgrounds or candidates with a pattern of 1–2 year stints
Consulting backgrounds are not disqualifying but will be scrutinized — be prepared to speak specifically to what you owned vs. what you touched
Data Mesh: proven experience transforming monolithic data warehouses into decentralized Data Mesh architectures including federated governance
Platforms: deep hands-on expertise with Databricks and Snowflake including compute optimization and cost management
Big Data: strong proficiency in Apache Spark, Delta Lake, Iceberg, and Hudi
Coding: expert-level Python, SQL, and Scala/Java nice to have
Streaming: Kafka or Kinesis; orchestration via Airflow, dbt, or Dagster
Governance: hands-on with Unity Catalog, Alation, or Collibra
MLOps: solid understanding of model registry, feature stores, and model serving

Nice To Haves

Coding: expert-level Python, SQL, and Scala/Java nice to have

Responsibilities

Lead the design and implementation of a scalable, decentralized Data Mesh architecture; define domain boundaries, data products, and federated governance standards
Drive the organization from centralized to decentralized data management — change management experience is as important as technical depth
Establish data contracts and self-service analytics capabilities across the organization
Architect and prototype resilient data pipelines using Databricks, Snowflake, and Spark, ensuring high availability and low latency
Drive adoption of open table formats (Delta Lake, Apache Iceberg) for ACID compliance, time travel, and schema evolution across the data lakehouse
Troubleshoot complex performance bottlenecks in distributed systems
Architect advanced data security patterns including dynamic data masking, tokenization, and row-level security
Implement centralized discovery and access control using Unity Catalog or equivalent enterprise data catalogs
Implement technical controls for data privacy regulations (GDPR, CCPA) including encryption at rest and in transit
Design and deploy high-throughput streaming architectures using Kafka or Pulsar for real-time data ingestion and processing
Deep understanding of workflow orchestration tools (Airflow, dbt, Dagster)
Build robust MLOps pipelines and feature stores bridging data engineering and production AI/ML
Collaborate with data scientists to operationalize machine learning models end-to-end
Mentor senior data engineers and architects; foster a culture of technical excellence and "Data as a Product" thinking
Treat data pipelines as code — strict CI/CD, unit testing, and version control practices expected