The opportunity
Crusoe is on a mission to accelerate the abundance of energy and intelligence . As the only vertically integrated AI infrastructure company built from the ground up, we own and operate each layer of the stack — from electrons to tokens — to power the world's most ambitious AI workloads.
What you'll do
Architecture Ownership: Own the end-to-end design of MDT: the job API and scheduler, the transfer workers on Kubernetes, storage integration and network paths. You decide what the system looks like and explain why.
Prototype Assessment and Migration: Evaluate the production prototype, decide what gets refactored, extended or replaced, and move its current users onto the managed service without interruption.
Throughput and Correctness: Set the data-plane strategy (sharding, concurrency, straggler handling, checksum verification) and the benchmarks that prove it, aiming for line-rate transfers over dedicated interconnects.
Cross-Team Contracts: Own the interfaces between MDT and Storage, Network, Kubernetes, IAM, Console, Billing and Monitoring. You define what partner teams build to and hold the line on it.
Extensibility and Build-versus-Buy: Design for new clouds and Crusoe-internal transfers with a pluggable transfer engine and regional worker pools, and evaluate open-source and vendor options before the team builds.
Operational Excellence: Set the bar for reliability, observability and on-call for every system the team ships.
What they're looking for
- Product Partnership: Work with Product on scope, pricing trade-offs and milestones, and surface edge cases early enough to change the plan.
- Technical Mentorship: Raise the team's level through design reviews, RFC feedback and direct coaching, so the team runs well without depending on you.
- Distributed Systems Depth: Hands-on experience designing and operating data-intensive distributed systems at scale, including the failure modes that only appear in production.
- Professional Experience: 8+ years shipping production software, including leading a service from prototype to general availability.
