Staff, MLOps Engineer

Sequen is hiring a Staff MLOps Engineer to build and operate the ML infrastructure powering their production ranking systems. This purely infrastructure-focused role sits at the intersection of machine learning, distributed systems, and platform performance, serving internal ML researchers. The role involves low-latency model serving, scaling pipelines, CI/CD for ML, observability, and evaluation loops. Candidates need deep production experience with PyTorch, Kubernetes, and high-throughput serving systems.

• Build and operate low-latency model deployment and distributed inference pipelines.
• Scale ranking systems from experimentation to production, optimizing latency, throughput, and cloud costs.
• Implement model CI/CD with versioning, canary releases, and zero-downtime rollbacks.
• Drive observability for real-time model performance, data drift, and system reliability.
• Develop evaluation loops to validate live inference accuracy and prevent training-serving skew.
• Collaborate with ML researchers and backend engineers on production serving topologies.

• 4–8+ years of experience in MLOps, ML engineering, or distributed infrastructure.
• Hands-on experience serving ultra-low-latency ML models under heavy concurrent workloads.
• Deep production-grade proficiency with Python and PyTorch.
• Comfort across major cloud platforms (AWS, GCP, or Azure) with Docker and Kubernetes.