5 months ago
Vancouver, CanadaMid Level / Senior
Responsibilities
- Support the infrastructure roadmap across DevOps and MLOps in partnership with engineering, data, ML, and product teams.
- Build and manage infrastructure for reliable production ML model deployment using CI/CD pipelines, APIs, and workflow orchestration tools.
- Automate model training, registry, validation, deployment, rollback, experiment tracking, and reproducible training workflows.
- Operate AWS and GCP training and inference infrastructure, BigQuery datasets, GPU workloads, Kubernetes clusters, and SageMaker environments.
- Monitor model performance, latency, data drift, resource usage, infrastructure reliability, and operational performance.
- Manage infrastructure as code with Terraform or CloudFormation and improve GitOps, deployment, observability, security, and automation practices.
- Create documentation, runbooks, alerting, incident reviews, and troubleshooting processes that help Product and Data teams operate systems independently.
- Evaluate cloud-native, DevOps, data infrastructure, and MLOps tools and recommend improvements to reliability, scalability, cost, and delivery speed.
Requirements
- 2–5 years of hands-on DevOps or cloud engineering experience in production environments.
- Expertise with Kubernetes, including EKS, Helm, and microservices.
- 1–2 years of AWS and Amazon SageMaker experience.
- 1–2 years of experience creating data pipelines.
- 1–2 years of experience with LLM deployments on Kubernetes.
- 1–2 years of experience with Kubernetes clusters using GPU-enabled nodes.
- Proven experience using Terraform for scalable, auditable infrastructure as code.
- Collaborative, solution-oriented approach with a passion for automation and continuous improvement.
Benefits
- Hybrid work arrangement, indicated by #LI-Hybrid.
- Fully remote candidates may be considered for select positions.
- Eligible permanent team members can participate in various benefits plans.
- Office locations include Boston, Vancouver, Chicago, and Vancouver, WA.
Tech Stack
Apache AirflowAWSFlaskGoogle BigQueryGoogle Cloud PlatformGrafanaHelmKubernetesMLflowPostmanPrometheusTerraform
Categories
About Later
Later builds an influencer marketing and social media platform for brands to discover creators, run campaigns, analyze performance, and manage creator payments, with optional managed services. The company also offers a link-in-bio tool used by marketers and creators. Founded in 2014 and headquartered in Boston, it is privately held and used by enterprise customers such as Nike, Wayfair, Unilever, and Southwest Airlines.