Senior MLOps Engineer - Later, Vancouver

About the role

Later is looking for a Senior MLOps Engineer to build and scale the infrastructure that takes Later's machine learning and AI systems from prototype to production. You will own the pipelines that train, retrain, deploy, and serve the models behind our products, including forecasting, recommendation, and LLM-based and agentic features, across AWS and GCP.

You will join the Infrastructure team alongside our DevOps engineers and work closely with Data Science, Data Engineering, Platform, and Product Engineering. This is an exciting opportunity for candidates searching for Later jobs and engineering jobs Vancouver who want to set the standard for MLOps.

Key duties

As part of the Later careers team, your responsibilities span strategy, architecture, building, operating, leadership, and collaboration:

• Own the MLOps roadmap with Infrastructure, Data Science, and Data Engineering leaders, aligned to product and AI growth plans.
• Evaluate and select tooling for orchestration, model registry, experiment tracking, LLM serving, and observability, and make clear build-or-buy recommendations.
• Shape the long-term architecture of our multi-cloud ML and data platform, with scalability, security, and cost built in from the start.
• Design, build, and operate infrastructure that deploys and scales ML models in production, using CI/CD, GitOps, Kubernetes, and orchestration tools such as Airflow, Kubeflow, or Argo Workflows.
• Own automated training and retraining pipelines, including model registry, validation, deployment, and rollback, and agree on the retraining cadence with Data Science.
• Deploy and operate LLM inference, including GPU capacity, cost controls, and performance monitoring.
• Manage training and inference infrastructure across AWS and GCP, including GPU workloads, using Terraform.
• Work with Data Engineering to source and prepare training data from BigQuery and S3, building pipelines and light transformations as needed.
• Build monitoring for model performance, latency, data drift, cost, and resource usage, using tools such as CloudWatch, Prometheus, and Grafana.
• Support model versioning, experiment tracking, and reproducible training workflows.
• Lead troubleshooting of complex production and deployment issues, and take part in the Infrastructure on-call rotation.
• Set and document standards for infrastructure-as-code, CI/CD, observability, secure deployment, and ML workflow automation.
• Write runbooks and self-serve guides so Product and Data teams can deploy, monitor, and troubleshoot with more autonomy.
• Lead incident reviews and post-mortems that improve how we operate.
• Mentor engineers through pairing, code review, and feedback.
• Explain technical tradeoffs clearly to both technical and non-technical stakeholders.

Qualifications

To qualify for this position among jobs in Vancouver, applicants should meet the following requirements:

• 7+ years of DevOps, cloud, or software engineering experience in production environments, including technical leadership: leading projects, setting direction, and mentoring others.
• Deep expertise with Kubernetes (EKS), Helm, and microservices, including running GPU workloads on Kubernetes.
• Hands-on AWS experience and a track record of using Terraform for scalable, auditable infrastructure as code.
• Experience deploying, serving, and scaling ML models in production, including CI/CD for ML.
• Experience building data and training pipelines in partnership with Data Science and Data Engineering teams.
• Strong system design skills, including designing ML systems end to end and explaining the tradeoffs.
• Clear communication with technical and non-technical audiences.

Benefits

Later provides competitive compensation and comprehensive perks for eligible team members:

• Market-based and data-driven approach to compensation, evaluated bi-annually.
• Salary range for candidates located in Canada of $110,000-160,000 CAD.
• Eligibility for all permanent team members to participate in various benefits plans as part of their overall compensation package.