You'd work across our distributed storage engine and the services that help customers store, manage, and protect their data.
This includes a fault-tolerant block layer that erasure-codes and distributes data beneath the file system, a transactional file system where aggregates roll up as they change so a query across billions of files answers in real time instead of walking a tree, and that same machinery drives quotas, capacity trends, snapshot policy, multiple access protocols, plus replication and backup - all running in userspace.
All of which makes Qumulo a distributed operating system problem: scheduling, memory hierarchy, protection, and concurrency. We're AI-forward and specific about what that means. Agents may write code, but engineers own every design decision, and nothing ships that isn't right.
Why work at Qumulo?
Qumulo blends the flexibility of a startup with the security of a mature and profitable organization. Here you’ll work closely with a small team of smart, capable engineers - and the folks you interview with are the ones you’ll work alongside. Opinions are expected, disagreement is normal, and you'd own designs from ideation to release. We value quality, correctness, and performance. Our architecture, protocols, and patents are public, so you can read how most of this works before you talk to us - and we’re small enough that your work shows up in the product.
As a Senior Member of Technical Staff, you will:
- Design and build across the file system and block layer beneath it, in C and Rust
- Work with a team of super smart people to solve hard problems
- Take designs, including your own, from rough thesis to shipped code
- Get real leverage out of coding agents in an environment where being close isn't good enough
- Debug across protocol, file system, cache, network, and object store
- Find the correctness bugs that only surface under real world conditions
- Leave everything you touch better than you found it
- Prototype the next thing worth building and bring it to the team
Our recent work includes federating subtrees of the namespace across regions and clouds, keeping search indexes coherent with a namespace that keeps changing, and getting data to GPU clusters before they ask for it.