Staff, MLOps Engineer

Sequen is an early-stage AI infrastructure startup seeking a Staff MLOps Engineer to build and scale the systems behind its production ranking models. This infrastructure-focused role involves low-latency model deployment, distributed inference pipelines, model CI/CD, and observability. The engineer will partner with ML research scientists to bring experimental models into high-throughput production environments.

• Build ML infrastructure: design, operate, and maintain low-latency model deployment, distributed inference pipelines, and real-time telemetry.
• Scale ranking systems: move models from experimentation to production, optimizing latency, GPU/CPU throughput, and cloud infrastructure costs.
• Implement model CI/CD: automated model versioning, canary releases, hot-swappable container rollouts, and zero-downtime rollbacks.
• Drive system observability: monitor real-time pipelines for model performance, data distribution drift, and reliability anomalies.
• Develop evaluation loops: robust evaluation pipelines to validate live inference accuracy and prevent training-serving skew.
• Optimize platform bottlenecks: eliminate performance bottlenecks across serving layers and improve model warm-up times.
• Collaborate with research: work with ML researchers and backend engineers to turn experimental models into resilient production systems.