Databricks
Principal Engineer, Compute Fleet Management
$264K–$322K
Bellevue, Washington +1 · Principal · Full-time · Posted Dec 11, 2025
- Location
- Bellevue, Washington
United States
About the job
Leads the design and operation of Databricks’ compute fleet systems to optimize cloud resource use, resilience, availability, and workload isolation across major cloud platforms.
What you'll do
- Set technical direction for provisioning, pooling, and optimizing large-scale cloud compute resources.
- Architect resilient, horizontally scalable systems for compute fleet management and platform bootstrapping.
- Lead strategic cross-team engineering initiatives and drive availability, utilization, and isolation outcomes.
What they ask for
- Hands-on experience developing and operating high-scale distributed systems on at least one major public cloud.
- Track record leading complex, cross-team, cross-layer, multi-quarter engineering initiatives from concept to execution.
- Proven ability to establish technical direction, build consensus, and lead across organizational boundaries.
- Strength in project planning, progress tracking, and managing complex cross-organizational dependencies.
Nice to have
- Experience managing and scaling a massive fleet of GPUs for AI/ML workloads.
- Experience developing and operating large-scale distributed systems across AWS, Azure, and GCP.
Pay
$264,300 — $322,300 USD (salary range)
Tools
AWS, Azure, GCP
AI focus
Preferred experience includes scaling GPU fleets for AI/ML workloads; the role otherwise focuses on compute infrastructure.
Model-written summary · the original posting is authoritative