Senior Software Engineer, Stateful Systems

About the work

Lattice is the software platform behind Anduril's autonomous systems. The Stateful Systems team owns its data layer: the time-series and storage systems that hold every track, sensor reading, and tasking record the platform produces. This runs in the field on real deployments, so performance and reliability here directly affect whether operators can do their jobs.

The team is small and the current focus is a rebuild of the core systems. It's a good time to join if you want a large surface area and real ownership.

What you'll do

• Time-Series: We're redesigning the time-series engine from the ground up to scale horizontally across many writers and nodes. The near-term requirement is concrete: 350+ sensor towers and tens of thousands of camera feeds, ingesting continuously, with both live and historical queries.
• Distributed storage. High-availability durability (WAL replication, leader election), a pluggable storage-engine layer, and a zero-downtime path to migrate between backends. 
• Correctness under partition. These systems sometimes run disconnected. Getting consistency and convergence right when the network is unreliable is a central part of the job.
• Hard distributed-systems problems. Consistency models, replication protocols, compaction and deletion at scale, back pressure, and durability guarantees.

What we're looking for

Required:

• Bachelor's degree in Computer Science, Engineering, or relevant field, or equivalent experience
• 6+ years of professional software development experience, with a focus on distributed systems or robotics
• Deep experience with Python, Go lang, C++, Rust, or similar systems programming languages
• Solid distributed-systems fundamentals. You can reason about replication, consistency, partition behavior, and failure modes, and explain why a given design does or doesn't converge.
• Production experience building or operating storage engines, databases, or large-scale data infrastructure.
• Comfort in a systems language. Depth matters more than the specific language you've used.
• A bias toward shipping and owning your work. The team is small and what you build is load-bearing.

Nice to have:

• Time-series systems, LSM or B-tree storage internals, or experience with FoundationDB, RocksDB, Pebble, or similar KV stores.
• Experience with systems that run in resource-constrained or intermittently-connected environments.

Why join

You get the ownership and pace of an early team, with a redesign in front of you and a lot of latitude to shape the architecture, backed by a real mission and the resources to do it well. The team is small enough that your work is visible and matters, and we put effort into supporting the people on it.