Software Engineer III, Machine Learning Platform

What you'll do

• Design, build, and maintain end-to-end machine learning training platforms.
• Optimize GPU training workloads to improve throughput, utilization, and reproducibility across single-node and distributed systems.
• Build and operate training infrastructure on Kubernetes platforms including resource management and troubleshooting.
• Enable Gen AI and LLM fine-tuning workflows with integrated evaluation harnesses and artifact governance.
• Implement observability for training systems through metrics, logs, dashboards, and operational runbooks.
• Improve developer experience via standardized containers, CI/CD pipelines, and self-service workflows.
• Optimize training costs and performance through right-sizing, scheduling policies, and infrastructure tuning.

What we're looking for

• Formal training or certification in software engineering concepts and 3+ years of applied experience.
• Experience running ML training in cloud environments and debugging issues across infrastructure and code.
• Strong Python skills with solid engineering practices like testing, code reviews, modular design, and dependency management.
• Experience building automation/CI for ML codebases including build, test, release, and deployment workflows.
• Hands-on experience with deep learning training workflows and at least one major framework such as PyTorch or TensorFlow.
• Experience with distributed training concepts (e.g., DDP/FSDP/DeepSpeed) and profiling/optimizing training systems for CPU/GPU utilization.
• Experience with Kubernetes fundamentals for compute-intensive workloads and AWS services like EKS, S3, IAM, and CloudWatch.
• Hands-on experience using enterprise-authorized AI-assisted software development tools to generate, validate, and refine code.