Role Summary
CrowdStrike is looking for an Engineer III - Data + ML Platform to join our infrastructure engineering team in Bengaluru under a hybrid model. In this role, you will own the reliability, performance profiling, and operational stability of the mission-critical ML platforms that power our Falcon cybersecurity ecosystem—processing over 3 trillion events each day. You will debug large-scale distributed computing frameworks (Ray, Spark, Kubernetes), troubleshoot GPU scheduling bottlenecks, build deep observability tooling, and lead incident post-mortems across our training and inference clusters.
Key Responsibilities
- Platform Reliability & Debugging: Diagnose root causes for production failures, kernel crashes, memory leaks, and distributed scheduling bottlenecks across Ray, Apache Spark, and Kubernetes.
- Distributed Compute Optimization: Profile and optimize high-throughput distributed ML workloads across Kubernetes and public cloud environments (AWS, GCP, Azure, or OCI), maximizing GPU and HPC cluster efficiency.
- Workflow & Tooling Maintenance: Troubleshoot and scale core ML pipeline components, including JupyterHub, SLURM, MLflow, Kubeflow, and Apache Airflow.
- Observability & SRE Operations: Architect distributed tracing, real-time health checks, and alerting runbooks with Prometheus and Grafana to maintain stringent latency, MTTR, and SLA targets.
Key Qualifications
- Education: Bachelor's or Master's degree in Computer Science, Software Engineering, or a related technical discipline.
- Experience: 5+ years of hands-on experience in distributed systems engineering and debugging ML platforms in production.
- Core Stack: Advanced Python debugging in Linux/Unix environments, deep production troubleshooting with Ray or Apache Spark, and container orchestration using Kubernetes and Docker.
- Nice to Have: Hands-on experience with SLURM cluster management, CUDA/GPU profiling, chaos testing, open-source ML platform contributions, and large-scale model inference pipelines.
- Professional Competencies: Disciplined operational triage, blameless post-mortem leadership, proactive mentoring, and an automation-first mindset.
About CrowdStrike
CrowdStrike is a global cybersecurity pioneer that redefined endpoint and cloud security with the world’s most advanced cloud-native, AI-driven platform built to stop breaches.