About this position:
We're looking for a Machine Learning Infrastructure Engineer to join our growing Data & Platform team and build the foundation that powers our AI and machine learning capabilities across Later's product portfolio. As our first dedicated ML Infrastructure Engineer, you will own the systems that support model experimentation, training, deployment, and monitoring at scale.
This role is critical to accelerating our data science initiatives and enabling future AI innovation. You'll design and operate reliable, secure, and scalable ML infrastructure that empowers data scientists and engineers to ship high-impact models with confidence. If you're excited about building robust ML systems in a fast-moving environment-and want to define the standard for ML Ops at Later-this is your opportunity.
What you'll be doing:Strategy
- Define and own the long-term ML infrastructure roadmap, ensuring it supports both current experimentation needs and future AI initiatives.
- Establish best practices for model lifecycle management, deployment standards, monitoring, and governance.
- Identify infrastructure gaps and proactively design scalable solutions to enable high-velocity ML development.
- Contribute to cross-functional technical planning, ensuring ML systems align with product and platform strategy.
Technical/ Execution
- Design, build, and maintain production-grade model deployment and inference systems using CI/CD pipelines, containerized services (Docker), and API frameworks (e.g., Flask).
- Automate end-to-end ML lifecycle workflows including training pipelines, model validation, registry management, deployment, and rollback strategies.
- Implement robust monitoring systems for model performance, latency, drift detection, and infrastructure health using tools such as CloudWatch, Prometheus, and Grafana.
- Operate across AWS and GCP environments to manage training and inference workloads, including GPU-based infrastructure and BigQuery datasets.
- Develop and maintain infrastructure-as-code (Terraform, CloudFormation) to ensure scalable, repeatable, and secure cloud environments.
- Implement and optimize CI/CD workflows (e.g., GitHub Actions, GitLab CI, Bitbucket Pipelines) for ML and infrastructure automation.
Team / Collaboration
- Partner closely with Data Scientists, Analysts, Platform Engineers, and Product Engineers to support end-to-end ML workflows.
- Translate data science experimentation needs into production-ready infrastructure solutions.
- Serve as the technical bridge between ML experimentation and productized deployment.
- Share knowledge and best practices to elevate ML maturity across teams.
Research/Best Practices
- Stay current on emerging ML Ops practices, tools, and frameworks to continuously improve system reliability and efficiency.
- Evaluate and implement model-serving frameworks (e.g., TorchServe, Seldon, TensorRT) where appropriate.
- Contribute to governance, reproducibility, and auditability standards for ML systems.
- Experiment with new tooling and workflows to improve reproducibility, performance, and developer velocity.
What success looks like:
- ML models move from experimentation to production quickly and reliably, with minimal manual intervention.
- CI/CD pipelines enable safe, repeatable deployments with clear rollback strategies.
- Model performance, drift, and infrastructure health are proactively monitored and observable.
- Infrastructure supports scalable GPU training and real-time inference without bottlenecks.
- Data scientists report improved velocity, reproducibility, and confidence in deploying models.
- ML systems are secure, compliant, and aligned with evolving product and AI strategy.
What you bring:
- 4+ years of experience in ML Ops, ML infrastructure, backend engineering, or related roles supporting production ML systems.
- Experience working in cloud-native environments (AWS and/or GCP) with hands-on deployment of ML workloads.
- Proven track record designing and implementing CI/CD pipelines for ML systems.
- Strong experience with Amazon SageMaker, Docker, Flask-based APIs, and infrastructure automation tools.
- Hands-on experience with ML lifecycle tooling such as MLflow, SageMaker Studio, or Weights & Biases.
- Experience managing container orchestration platforms (Kubernetes, EKS, or GKE).
- Strong programming experience in Python (additional experience in Go, Java, or Scala is a plus).
- Experience working with infrastructure-as-code tools such as Terraform or CloudFormation.
- Familiarity with observability tools such as CloudWatch, Prometheus, Grafana, Datadog, or centralized logging platforms.
- Experience managing GPU-based workloads and scaling training/inference systems.
- Familiarity with data infrastructure tools such as BigQuery and cloud-native data pipelines.
- Bonus: Experience supporting LLMs or generative AI pipelines, distributed training systems, feature stores (e.g., Feast), real-time inference systems, or ML governance fram