Diagnose and resolve issues across Ray, Spark, Airflow, MLflow, JupyterHub, Kubeflow, and SLURM
Perform root cause analysis on production incidents affecting training and inference pipelines
Debug performance bottlenecks, resource contention, memory leaks, and scheduling conflicts
Develop debugging tools and diagnostic frameworks
Profile and optimize Ray clusters and Spark jobs on K8s and Cloud (EMR/Dataproc)
Troubleshoot JupyterHub spawner issues, kernel crashes, and resource allocation
Optimize SLURM job scheduling, GPU allocation, and HPC cluster utilization
Build observability solutions and automated health checks
Develop runbooks, alerting workflows, and incident response procedures
Maintain platform stability metrics (SLAs, error rates, latency)
Partner with ML and ML Platform engineers to resolve workflow issues
Conduct post-mortems and mentor on debugging techniques

12+ years in distributed systems engineering
5+ years debugging ML platforms in production
Deep expertise in 3+ one of: Ray, Spark, JupyterHub, SLURM, K8
Performance profiling, optimization, and capacity planning
Technical Skills (Expertise in at least one):
Distributed ML: Ray, Spark, SLURM, Jupyter Ecosystem (debugging failures, performance tuning)
ML Platforms: Airflow, MLflow, JupyterHub (troubleshooting core components)
Infrastructure: Kubernetes, Docker, AWS/GCP/Azure/OCI
Observability: Profiling tools, distributed tracing, Prometheus, Grafana, log aggregation
Programming: Expert Python debugging, multi-language proficiency, Linux/Unix

Market leader in compensation and equity awards
Comprehensive physical and mental wellness programs
Competitive vacation and holidays for recharge
Paid parental and adoption leaves
Professional development opportunities for all employees regardless of level or role
Employee Networks, geographic neighborhood groups, and volunteer opportunities to build connections
Vibrant office culture with world class amenities
Great Place to Work Certified™ across the globe

Senior ML Platform Engineer

Key skills