MTS - AI / MLOps

THE ROLE

Join the Pure Solutions team as a Senior MLOps Solutions Engineer to architect and build high-scale, enterprise-grade AI/ML solutions. You will be instrumental in integrating Pure Storage platforms with the evolving open-source MLOps ecosystem (Kubeflow, MLflow, Ray) to operationalize the complete machine learning lifecycle. This role requires a creative technologist with deep Python expertise to drive innovation and enable our customers and partners to achieve production AI success.

WHAT YOU'LL DO

• Design and Automate MLOps Pipelines: Lead the development of end-to-end MLOps workflows using CI/CD tools (Git/Jenkins) and orchestration platforms (MLflow/Kubeflow), specifically integrating Pure Storage's FlashBlade, FlashArray, and Portworx as the high-performance data plane for data ingestion, training, and inference.

• Build High-Performance AI/ML Reference Architectures: Create validated, repeatable deployment models using Infrastructure as Code (e.g., Ansible, Terraform) for AI/ML environments spanning bare metal, virtual machines, and GPU-accelerated Kubernetes clusters, ensuring optimal performance for distributed training.

• Optimize and Operationalize GPU Inference: Architect and implement solutions for high-throughput, low-latency model serving, utilizing technologies like NVIDIA Triton Inference Server and advanced optimization techniques (quantization, model sharding like DeepSpeed/Megatron-LM, and dynamic batching) for large models (LLMs).

• Enable Sales and Drive Ecosystem Adoption: Develop automated, GPU-enabled MLOps lab environments, high-quality technical documentation, and live demonstrations to enable global sales, field engineering teams, and strategic partners on new AI integrations, directly impacting solution adoption and revenue.

• Define Strategic MLOps Direction: Collaborate closely with Data Scientists and Product Management to influence the technical strategy for AI platform integrations and provide essential input into future product roadmaps related to accelerated compute and high-performance storage requirements.

WHAT YOU BRING

• Deep MLOps Pipeline & Infrastructure as Code (IaC) Expertise: Hands-on experience designing, building, and automating MLOps workflows using orchestration tools (e.g., Kubeflow, MLflow, Vertex AI, SageMaker) and proficiency with IaC tools such as Terraform or Ansible.

• Advanced Python & Deep Learning Framework Proficiency: Expert-level skills in Python for Data Science and MLOps, including libraries like pandas and NumPy, and demonstrated experience with Deep Learning frameworks, particularly PyTorch, focusing on distributed training and model handling.

• Expertise in GPU-Accelerated Computing & Container Orchestration: Strong practical knowledge of GPU computing principles (CUDA), technologies like NVIDIA Triton Inference Server, and expert-level working knowledge of Kubernetes for GPU resource management, scheduling, and persistent storage for containers.

• High-Performance Storage and Data Center Understanding: Solid comprehension of high-performance data center infrastructure, including high-speed networking (e.g., RoCE, 100GbE), and storage platforms optimized for high-throughput, low-latency AI/ML workloads (e.g., high-performance S3, parallel file systems).

• We are primarily an in-office environment and therefore, you will be expected to work from the {{OFFICE_LOCATION}} office in compliance with Pures policies, unless you are on PTO, or work travel, or other approved leave.