Mercury is hiring a remote Senior ML Operations Engineer!
Mercury is expanding machine learning’s role in risk decisioning, and its Machine Learning Platform team is building a paved path from trained model to reliable production. This senior remote role joins a new platform group focused on real-time fraud and financial-crime scoring. You would own the inference service and deployment lifecycle, treating low latency and high availability as core design constraints while giving data scientists a self-serve route to register, ship, and monitor models. The work spans observability, drift-triggered retraining, experimentation, and explainability. Mercury expects strong product ownership and wants someone eager to help shape a brand-new platform team.
Key responsibilities
- Build and run the real-time inference service that scores models for the risk decision engine, with low latency and high availability as first-class requirements.
- Own model deployment infrastructure, including registry and versioning, CI/CD with performance, bias, and consistency checks, shadow mode, and staged rollouts.
- Create model observability covering availability, latency, errors, and drift detection that can trigger retraining.
- Partner with Risk Data Science to move models from development handoff into production operation under platform ownership.
- Implement experimentation patterns such as champion/challenger and canary routing, plus explainability outputs like SHAP attributions.
- Take active product ownership and help shape a new platform team, self-organizing on small and medium projects.
Key requirements
- 5+ years in machine learning engineering, backend software engineering, MLOps, or a closely related field.
- Production experience deploying, serving, and operating ML models in low-latency, high-availability environments.
- Strong Python backend fundamentals and experience with API frameworks such as FastAPI or Flask.
- Hands-on experience with model lifecycle tooling: registries, CI/CD for models, versioning, and staged rollout patterns like shadow, canary, and champion/challenger.
- Experience building observability and alerting for production services, including model-specific signals such as drift.
- Comfort with the data layer ML depends on: SQL, key-value or low-latency stores such as Redis or DynamoDB, and streaming pipelines such as Kafka, Kinesis, or Redpanda.
- Nice to have: modern data stack exposure (Snowflake, dbt, Dagster, Airflow), regulated or audit-sensitive environments, and willingness to work across Haskell, React, and TypeScript.
- Seniority: Senior
- Work mode: Remote
- Salary: CAD $157,400–$196,800; US $166,600–$208,300
- Key skills: Python, FastAPI, Flask, model registries, CI/CD, shadow deployment, canary rollout, champion/challenger, observability, drift detection, SQL, Redis, DynamoDB, Kafka, Kinesis
Want more jobs?
Join our mailing list to receive updates about new data science and Senior ML Operations Engineer jobs.
