Staff Software Engineer - AI Inference Platform

CoreWeave; Warszawa; On-site; PLN 369,000 - 493,000; Full time

Benefits offered by this job

Family-level Medical Insurance

Family-level Dental Insurance

Generous Pension Contribution

Life Assurance at 4x Salary

Critical Illness Cover

Employee Assistance Programme

Tuition Reimbursement

Work culture focused on innovation

Job summary

CoreWeave is seeking a Staff Software Engineer for the Inference team in Warsaw to lead architecture, performance, and reliability of a Kubernetes-native inference platform powering AI workloads. You will guide cross-team initiatives, optimize for latency, and drive cost-per-token improvements with advanced techniques.

You will implement optimisations, build benchmarking frameworks, and mentor engineers, while shaping high-impact engineering practices and observability across CoreWeave.

Qualifications

• Bachelor's degree in CS/engineering or equivalent practical experience.
• 8+ years building large-scale distributed systems or cloud platforms.
• Strong coding in Go, Python, or C++.
• Deep Kubernetes production experience.
• Strong understanding of networked systems and performance optimization.
• Hands-on with inference systems including batching and caching.
• Experience improving tail latency and reliability with metrics-driven engineering.

Responsibilities

• Define and drive architecture across multiple teams and services.
• Improve performance, reliability, and observability of the Kubernetes-based inference platform.
• Lead cross-cutting design initiatives (routing, scheduling, resource management, cost-per-token).
• Implement advanced inference optimisations (speculative decoding, KV-cache reuse).
• Establish benchmarking frameworks and cross-functional alignment across infra boundaries.

Skills

Go

Python

C++

Kubernetes

Distributed systems

Networking

Inference systems

SLIs/SLOs

Autoscaling

Education

Bachelor's degree in CS/Engineering

Tools

CUDA

NCCL

RDMA

Job description

CoreWeave is seeking a Staff Software Engineer for the Inference team in Warsaw to lead architecture, performance, and reliability of a Kubernetes-native inference platform powering AI workloads. You will guide cross-team initiatives, optimize for latency, and drive cost-per-token improvements with advanced techniques.

You will implement optimisations, build benchmarking frameworks, and mentor engineers, while shaping high-impact engineering practices and observability across CoreWeave.