MLops Engineer

Arrayo; Massachusetts; On-site; USD 130,000 - 185,000; Full time

Job summary

Arrayo is seeking an MLops Engineer to lead the scaling of machine learning training pipelines and ensure robust end-to-end workflows. The role focuses on Flyte, GPU-optimized Kubernetes, Docker, and distributed training frameworks like Ray to optimize ML infrastructure.

You will orchestrate workflows, scale multi-node GPU training, and collaborate with data scientists to productize experiments while ensuring reproducibility and cost efficiency.

Qualifications

• Experience with GPU scheduling on Kubernetes and cluster autoscaling.
• Hands-on with Flyte or similar tools (Airflow, Prefect).
• Deep knowledge of distributed ML training (e.g., PyTorch DDP, Ray, Horovod).

Responsibilities

• Develop and maintain ML workflows using Flyte to manage training, testing, and deployment.
• Scale large-scale ML training systems on GPU-backed Kubernetes clusters with auto-scaling and tuning.
• Implement distributed model training pipelines using Ray for parallelization and efficiency.
• Design, build, and optimize Docker images for ML workloads with reproducibility and security.
• Debug and optimize GPU utilization, memory, and compute bottlenecks during training and inference.
• Integrate monitoring for ML jobs, track resource consumption, and enforce cost-efficient resource usage.
• Collaborate with data scientists and ML engineers to productize and scale ML experiments.

Skills

Workflow orchestration

Distributed computing

GPU optimization

ML training

Collaboration with data scientists

Tools

Kubernetes

Flyte

Docker

Ray

PyTorch DDP

Horovod

CUDA

NCCL

GitHub Actions

ArgoCD

Prometheus

Grafana

Job description

MLops Engineer (Training Scalability & Workflow Optimization)

Overview

We are seeking an MLops Engineer to lead the scaling of machine learning training pipelines and ensure the robustness and efficiency of our end-to-end ML workflows. This role focuses on leveraging Flyte, Kubernetes (GPU optimization), Docker, and distributed training frameworks such as Ray to optimize and streamline our ML infrastructure.

Responsibilities

• Workflow Orchestration: Develop and maintain ML workflows using Flyte to manage complex ML pipelines for training, testing, and deployment.
• Training Scalability: Architect and scale large-scale ML training systems on GPU-backed Kubernetes clusters, including auto-scaling and performance tuning for multi-node/multi-GPU workloads.
• Distributed Computing: Implement distributed model training pipelines using frameworks like Ray for parallelization and resource efficiency.
• Containerization: Design, build, and optimize Docker images for ML workloads with a focus on reproducibility and security.
• Resource Optimization: Debug and optimize GPU utilization, memory, and compute bottlenecks during training and inference phases.
• Monitoring & Maintenance: Integrate monitoring for ML jobs, track resource consumption, and enforce cost-efficient resource utilization.
• Collaboration: Work closely with data scientists and ML engineers to productize and scale ML experiments.

Qualifications

• Strong proficiency with Kubernetes (GPU scheduling, Helm, cluster autoscaling).
• Hands‑on experience with Flyte or similar workflow orchestration tools (Airflow, Prefect).
• Deep knowledge of distributed ML training (e.g., PyTorch DDP, Ray, Horovod).
• Expertise in Docker and container lifecycle management.
• Solid understanding of GPU hardware/software stack (CUDA, NCCL).
• Familiarity with CI/CD for ML (MLops pipelines using tools like GitHub Actions, ArgoCD).
• Bonus: Familiarity with observability tools for ML systems (Prometheus, Grafana).