Senior MLOps Engineer - SRE | DevOps

This position is listed on behalf of a partner company, who manages all applications and next steps. Our partner is looking for a Senior MLOps Engineer - SRE | DevOps based in Brazil.

This role sits at the core of a modern AI/ML platform, responsible for ensuring that machine learning and LLM systems run reliably, efficiently, and at scale in production environments.
You will operate at the intersection of infrastructure, DevOps, and machine learning, owning the lifecycle from model deployment to production-grade inference services.
The position involves solving complex engineering challenges such as latency optimization, autoscaling, cost efficiency, and reliability of AI workloads.
You will help define how ML systems are built and operated, introducing strong SRE practices to environments that demand high availability and performance.
A key part of your work will involve designing scalable, automated, and observable ML pipelines and infrastructure on cloud-native platforms.
This is a high-impact role for a senior engineer who thrives in deep technical ownership and wants to shape the future of AI infrastructure.

• Design, build, and operate scalable ML and inference infrastructure supporting real-time and batch workloads across multiple tenants.
• Own the end-to-end ML deployment lifecycle, including model registry, versioning, rollout strategies (canary, A/B, shadow), and safe rollback mechanisms.
• Operate and optimize production-grade AI and LLM workloads, managing inference providers, throttling, quotas, and fallback strategies under load.
• Develop and maintain reproducible ML pipelines for training, evaluation, and deployment with full lineage and automation.
• Implement Infrastructure-as-Code practices using Terraform, ensuring scalable multi-account cloud architectures.
• Manage GitOps workflows using tools such as ArgoCD to ensure reliable and consistent deployments across environments.
• Operate Kubernetes-based infrastructure (AWS EKS), including GPU scheduling, workload isolation, and cost-aware scaling strategies.
• Define and enforce SRE best practices, including SLOs, observability, incident response, and performance monitoring for ML systems.
• Drive cost optimization initiatives across ML workloads, including resource right-sizing and efficient infrastructure utilization.
• Improve automation across the ML lifecycle using modern engineering and agentic coding tools.

Requirements:

• 5+ years of experience in Platform Engineering, SRE, DevOps, or MLOps roles, operating production systems at scale.
• Strong hands-on experience deploying and managing ML/AI workloads in production environments.
• Deep SRE expertise, including SLO definition, incident response, postmortems, and reliability engineering practices.
• Advanced experience with Infrastructure-as-Code using Terraform in complex, multi-account environments.
• Strong GitOps experience with declarative infrastructure and deployment workflows.
• Deep expertise in Kubernetes, including production operations and failure-mode troubleshooting.
• Strong AWS knowledge, including networking, IAM, compute, storage, and distributed architectures.
• Experience building CI/CD pipelines using tools such as GitHub Actions, GitLab CI, CircleCI, or similar.
• Strong automation mindset with ability to eliminate manual operational work through engineering solutions.
• Familiarity with agentic coding tools and ability to use them effectively in infrastructure and pipeline development.
• Strong communication skills to articulate technical decisions, trade-offs, and incident analysis clearly.

Nice to have:

• Experience with GPU/accelerator scheduling and node lifecycle management (e.g., Karpenter).
• Experience operating LLM inference systems at scale, including quota management, caching, and guardrails (e.g., AWS Bedrock or similar).
• Experience with ML orchestration tools such as Argo Workflows, Kubeflow, Airflow, or SageMaker Pipelines.
• Familiarity with ML observability tools, drift detection, and model monitoring practices.
• Background in FinOps and cost attribution for large-scale inference systems.
• Experience with multi-tenant infrastructure and isolation strategies.
• Exposure to feature stores, model registries, and experiment tracking tools such as MLflow or Feast.
• Experience scaling ML platforms in high-growth or startup-to-enterprise environments.

Benefits:

• Fully remote work model with flexibility.
• Opportunity to work on cutting-edge AI and ML infrastructure at scale.
• High ownership environment with direct impact on platform architecture and evolution.
• Exposure to modern cloud-native technologies, Kubernetes, and distributed systems at production scale.
• Collaborative engineering culture focused on automation, reliability, and innovation.
• Work aligned with global time zones (EST/PST) for structured collaboration.
• Continuous technical challenges involving LLMs, ML systems, and large-scale infrastructure.
• Strong emphasis on engineering autonomy and senior-level decision-making.