Principal Software Engineer 2 - OLTP

Snowflake•Menlo Park, CA

3d•$304,000 - $380,000

About The Position

Snowflake is building the next generation of transactional systems for applications, AI and agents. Across Snowflake Postgres and Unistore, we are combining the performance, reliability and developer experience expected from modern operational databases with Snowflake's cloud-native architecture and governed data platform. We're looking for a Principal Engineer 2 to help define and drive the architecture across database internals, storage, compute and cloud infrastructure. A core part of the role is designing transactional systems for agent-driven workloads, where isolated state, rapid database lifecycle, safe autonomous operations, and deep integration with enterprise data are first-class requirements.

Requirements

15+ years of experience building databases, distributed systems, storage systems or large-scale cloud infrastructure, with sustained impact at significant technical scope.
Deep expertise in OLTP and database systems, including areas such as transactions, concurrency control, storage engines, indexing, query processing, replication, recovery and high availability.
Strong understanding of cloud infrastructure for low-latency stateful systems, including compute, storage, networking, resource isolation and failure handling.
Experience designing highly elastic, multi-tenant, or rapidly provisioned stateful systems where isolation and efficient lifecycle management are first-class concerns.
Strong instincts for how database architecture should evolve for AI-native applications and autonomous agents, including ephemeral state, sandboxing, safe transactional actions and programmatic database lifecycle.
Experience making difficult architectural tradeoffs across latency, throughput, consistency, availability, durability, scalability, isolation and cost.
A track record of taking technically ambitious systems from architecture through production at meaningful scale.
Experience influencing technical direction across multiple teams or a broad product area.
Strong systems engineering instincts and the ability to work across abstractions, from database internals to distributed infrastructure.
Deep PostgreSQL expertise is a strong advantage, but experience building other sophisticated transactional or distributed database systems is equally relevant.
Strong systems programming experience; C, C++, or Rust experience is a plus.
Exceptional technical judgment and communication skills, including the ability to influence engineers, executives and customers.

Responsibilities

Define and drive the long-term technical architecture for OLTP across Snowflake Postgres, Unistore and the infrastructure they depend on.
Lead major architectural decisions across transactions, storage, concurrency control, replication, recovery, query execution and database scalability.
Shape the compute and storage architecture for low-latency, highly available transactional systems, including resource isolation, elasticity, networking, caching and container infrastructure.
Build database primitives for agent-driven workloads, including fast creation of isolated transactional environments, checkpointing and rollback, and efficient lifecycle management.
Define how agents safely read and modify operational state, with strong guarantees around consistency, recovery, access control and governance.
Determine where PostgreSQL architecture and compatibility provide leverage and where alternative approaches can deliver meaningful advantages.
Design for workloads where databases and transactional environments are created and operated programmatically at machine speed rather than primarily through human administration.
Lead complex, cross-organizational engineering efforts and establish technical direction across multiple teams.
Work closely with product leaders, senior engineering leadership, and strategic customers on the most important technical and product decisions.
Raise the technical bar through architecture reviews, mentoring and hands-on leadership on the hardest systems problems.