Senior Distributed Caching Engineer — High-Scale Storage

DoorDash; Seattle (WA); Hybrid; USD 160,000 - 235,000; Full time

Benefits offered by this job

401(k) matching

Paid parental leave

Healthcare, dental, vision

Paid time off & holidays

Job summary

DoorDash’s Storage organization manages stateful online systems powering large-scale caching for DoorDash, Wolt, and Deliveroo. The Distributed Caching team handles ElastiCache, Boulder, and smart clients, delivering reliable performance across major markets.

This hybrid Seattle-based role focuses on scaling Boulder, re-platforming the Distributed Lock Service, and building self-serve tooling for workload-to-backend mapping, enabling customers to describe workloads and get the right backend

Qualifications

• 6+ years of full-time experience designing, building, and operating distributed, stateful systems in production.
• Proficient in Go, or Java/Kotlin with concurrency and backend services.
• Hands-on experience with distributed caching technologies (Redis/Valkey, Memcached, or similar).
• Understanding of consistency, sharding, replication, and consensus in distributed systems.

Responsibilities

• Own high-leverage caching and locking services with measurable customer impact.
• Build and scale Boulder, our KVRocks-backed key-value store, including compaction and sharding.
• Re-platform the Distributed Lock Service onto a strongly consistent backend with sharding.
• Contribute to the standardized smart caching client with failover, retry, and observability defaults.

Skills

Go

Java/Kotlin

Distributed systems

Caching

KVRocks

Kubernetes

AI tools

Tools

Redis/Valkey

ElastiCache

Memcached

Valkey-go

TiKV

Job description

DoorDash’s Storage organization manages stateful online systems powering large-scale caching for DoorDash, Wolt, and Deliveroo. The Distributed Caching team handles ElastiCache, Boulder, and smart clients, delivering reliable performance across major markets.

This hybrid Seattle-based role focuses on scaling Boulder, re-platforming the Distributed Lock Service, and building self-serve tooling for workload-to-backend mapping, enabling customers to describe workloads and get the right backend