Senior Platform Engineer: Kubernetes & Reliability

Coreweave; United States; On-site; USD 180,000 - 300,000; Full time

Benefits offered by this job

Medical Insurance

Life Insurance

Disability Insurance

401(k) with employer match

Tuition Reimbursement

ESPP

Flexible PTO

Catered lunch in office

Mental wellness benefits

Flexible childcare support

Job summary

CoreWeave in the United States seeks senior engineers to lead designs, raise engineering standards, and improve latency, throughput, and reliability across our Kubernetes-native inference platform. You will partner with product, orchestration, and hardware teams to meet strict P99 SLAs at scale.

You will mentor IC1/IC2 engineers; for IC4 you’ll own multi-service areas such as routing & adaptive scheduling, cost-per-token analytics, and GPU resource isolation.

Qualifications

• Experience building distributed systems or cloud services at scale.
• Strong coding in Python or Go and familiarity with networked systems.
• Production Kubernetes experience, CI/CD, and observability stacks.

Responsibilities

• Lead design reviews and drive architecture; decompose multi-service work into milestones.
• Define and own SLIs/SLOs; ensure reliability improves release-to-release.
• Implement optimizations and quantify impact on latency and throughput.
• Strengthen incident posture with capacity planning and autoscaling strategies.
• Mentor IC1/IC2 engineers and review cross-team designs.
• For IC4: own areas spanning multiple services like routing and adaptive scheduling.

Skills

Python

Go

Kubernetes

CI/CD

Observability

Education

Bachelor's or Master’s in CS/EE

Tools

CUDA kernels

NCCL/SHARP

RDMA/NUMA

GPU interconnect

Job description

CoreWeave in the United States seeks senior engineers to lead designs, raise engineering standards, and improve latency, throughput, and reliability across our Kubernetes-native inference platform. You will partner with product, orchestration, and hardware teams to meet strict P99 SLAs at scale.

You will mentor IC1/IC2 engineers; for IC4 you’ll own multi-service areas such as routing & adaptive scheduling, cost-per-token analytics, and GPU resource isolation.