Principal Software Engineer II - Interoperable Foundations

Snowflake Snowflake · Data AI · WA-Bellevue, United States · Engineering

This Principal Software Engineer role focuses on building the foundational data layer for AI and agentic innovation at Snowflake. The core mission involves a 0->1 redesign of metadata systems, Apache Iceberg, and AI-ready storage to support large-scale AI workloads and enable AI agents to navigate and act upon massive datasets. The role requires deep expertise in distributed systems, database internals, and cloud-scale architecture.

What you'd actually do

  1. Lead the ground-up redesign of our core Metadata systems, influencing the transaction frameworks that power query, DML, and AI-driven data interactions in addition to extending our lead on platform capabilities such as Zero Copy Cloning and Cross-Region / Cross-Cloud Replication.
  2. Drive industry-leading innovation on Apache Iceberg, ensuring it serves as a first-class citizen for Analytics, Data Engineering, and high-performance AI workloads.
  3. Build the foundational metadata capabilities - from richer metadata for AI workloads, to lifecycle management and global replication (Snowgrid)—that allow AI agents to navigate, understand, and act upon massive datasets in a secure and governed fashion.
  4. Solve "impossible" challenges in fault-tolerance, scalability, high availability, and performance for services that must meet the demands of a rapidly growing global ecosystem.
  5. Own the end-to-end architecture, influence the product roadmap, and identify the next innovations that will keep Snowflake at the cutting edge of the AI era.

Skills

Required

  • 15+ years of experience designing and supporting large-scale distributed services
  • Deep Database Internals Expertise: Comprehensive knowledge of metadata management, query optimization, clustering, compaction, catalogs, and large-scale data management
  • Systems Mastery: Excellent understanding of distributed systems, durability and correctness fundamentals, multi-threading, memory management, and storage performance
  • Cloud-Scale Track Record: A history of implementing creative, 0->1 solutions for complex problems at massive scale
  • CS Fundamentals: Strong mastery of data structures, algorithms, and the trade-offs between consistency and performance

Nice to have

  • advanced degree is preferred

What the JD emphasized

  • 0->1 redesign
  • 0->1 solutions

Other signals

  • foundational data layer for AI agents
  • accelerates agentic innovation
  • richer metadata for AI workloads
  • AI-ready storage