Platform Engineer, Writer

• You have professional experience with:
• Model training
• Huggingface Transformers
• Pytorch
• VLLM
• TensorRT
• Infrastructure as code tools like Terraform
• Scripting languages such as Python or Bash
• Cloud platforms such as Google Cloud, AWS or Azure
• Git and GitHub workflows
• Tracing and Monitoring
• Familiar with high-performance, large-scale ML systems
• You have a knack for troubleshooting complex systems and enjoy solving challenging problems
• Proactive in identifying problems, performance bottlenecks, and areas for improvement
• Take pride in building and operating scalable, reliable, secure systems
• Are comfortable with ambiguity and rapid change

Desirable

• As a Platform engineer, MLOps, you will be critical to deploying and managing cutting-edge infrastructure crucial for AI/ML operations, and you will collaborate with AI/ML engineers and researchers to develop a robust CI/CD pipeline that supports safe and reproducible experiments
• Your expertise will also extend to setting up and maintaining monitoring, logging, and alerting systems to oversee extensive training runs and client-facing APIs
• You will ensure that training environments are optimally available and efficiently managed across multiple clusters, enhancing our containerization and orchestration systems with advanced tools like Docker and Kubernetes
• This role demands a proactive approach to maintaining large Kubernetes clusters, optimizing system performance, and providing operational support for our suite of software solutions
• Work closely with AI/ML engineers and researchers to design and deploy a CI/CD pipeline that ensures safe and reproducible experiments
• Set up and manage monitoring, logging, and alerting systems for extensive training runs and client-facing APIs
• Ensure training environments are consistently available and prepared across multiple clusters
• Develop and manage containerization and orchestration systems utilizing tools such as Docker and Kubernetes
• Operate and oversee large Kubernetes clusters with GPU workloads
• Improve reliability, quality, and time-to-market of our suite of software solutions
• Measure and optimize system performance, with an eye toward pushing our capabilities forward, getting ahead of customer needs, and innovating for continual improvement
• Provide primary operational support and engineering for multiple large-scale distributed software applications