Overview
Senior MLOps Solutions Engineer to architect and build high-scale, enterprise-grade AI/ML solutions, integrating Pure Storage platforms with open-source MLOps tools.
What you'll do
- Design and automate end-to-end MLOps workflows using CI/CD and orchestration tools.
- Integrate Pure Storage platforms (FlashBlade, FlashArray, Portworx) into MLOps for ingestion, training, and inference.
- Build infrastructure-as-code reference architectures for bare metal, VMs, and GPU Kubernetes clusters.
- Optimize and operationalize GPU inference for high-throughput, low-latency model serving.
- Create solutions for large-model serving using NVIDIA Triton Inference Server and optimization techniques.
- Develop GPU-enabled MLOps lab environments, documentation, and live demos to support adoption.
- Partner with data scientists and product management to influence MLOps integration strategy and roadmaps.
What you'll need
- Experience designing, building, and automating MLOps workflows with tools like Kubeflow and MLflow.
- Proficiency with infrastructure as code tools such as Terraform or Ansible.
- Expert-level Python skills for data science and MLOps, including pandas and NumPy.
- Experience with deep learning frameworks, particularly PyTorch, for distributed training and model handling.
- Knowledge of GPU-accelerated computing (CUDA) and GPU inference serving with NVIDIA Triton Inference Server.
- Expert-level working knowledge of Kubernetes for GPU resource management, scheduling, and persistent storage.
- Understanding of high-performance data center infrastructure and storage/networking for AI/ML workloads.
Details
- Location: Bangalore, India.
- Primarily in-office, expected to work from the office location unless on PTO, travel, or other approved leave.
Read the full description and apply on the company’s own careers page.
