Prinicpal Software Engineer - Streaming Primitives

Data pipelines are foundational infrastructure - when they're fast, correct, and maintainable, customers build on them with confidence. If you've spent the bulk of your career building large-scale data infrastructure - designing streaming or transformation primitives, reasoning hard about consistency and fault tolerance, and owning the systems that run under millions of customer workloads - this role might be for you.

You'll be working on the streaming and transformation layer at Snowflake: the constructs that define how customers move, shape, and maintain data. AI has a real presence in this work - in how customers use these pipelines and in how we think about building them - but the core job is hard distributed systems engineering, and that's what we're hiring for.

We build the core data engineering primitives that power Snowflake's streaming and transformation capabilities. From the constructs customers use to define real-time pipelines to the execution fabric that makes those pipelines reliable and cost-efficient at cloud scale, our team owns the full stack of declarative data engineering.

We're a small, high-ownership team operating close to the product - which means your decisions ship, your architecture matters, and your fingerprints are on some of the most-used features in Snowflake's data engineering portfolio.

- Define and drive the technical direction for Snowflake's core data engineering and streaming transformation primitives, spanning Streams, Tasks, Dynamic Tables, and adjacent pipeline constructs.

- Identify and lead multi-quarter technical investments - performance, scalability, correctness, and reliability - translating ambiguous problem spaces into concrete engineering plans with measurable outcomes.

- Partner with product, research, and peer engineering teams to co-design primitives that compose cleanly across the data engineering stack.

- Operate as a force multiplier: run architectural reviews, set the technical bar for design documents, and help engineers grow through high-quality feedback and sponsorship.

- Work directly with customers and field teams to understand real-world usage patterns; use that signal to prioritize what matters next.

- Contribute to Snowflake's technical reputation - through internal design influence, external talks, or research publications in the data engineering space.

- 15+ years of experience designing, building, and operating large-scale distributed data systems.

- Deep expertise in at least one core area: stream processing, declarative query execution, pipeline orchestration, or data transformation at scale.

- Strong computer science fundamentals - distributed systems, algorithms, fault tolerance, and consistency models.

- Proficiency in C++ or Java; comfort with systems-level reasoning (latency, throughput, resource efficiency at cloud scale).

- Demonstrated ability to lead cross-team technical initiatives from blank-page architecture through production at petabyte scale across thousands of concurrent workloads.

- Strong written and verbal communication skills; ability to represent complex technical trade-offs clearly to engineering, product, and leadership audiences.

- Experience with a major analytical DBMS (Snowflake, BigQuery, Redshift, Databricks, Teradata).

- Hands-on background in streaming or event-driven systems (Flink, Kafka, Spark Structured Streaming).

- Familiarity with the broader data engineering ecosystem: dbt, Airflow, Fivetran, Iceberg, Delta Lake.

- Experience with CDC, change propagation, or incremental computation patterns.

- Advanced degree (MS or PhD) in Computer Science, with emphasis on database or distributed systems.