Executive Director Machine Learning Engineer, MLOps

What you'll do

• Build and maintain pipelines for distributed training on GPU-enabled clusters.
• Develop and manage high-volume real-time and batch inference systems.
• Implement quantization techniques and deploy open-weight LLMs using vLLM on Ray.
• Manage and optimize vector databases to support advanced AI applications.
• Establish comprehensive monitoring and observability pipelines for system health and performance.
• Implement fine-tuning and reinforcement learning algorithms on large compute clusters.
• Perform hyper-parameter tuning at scale and production validation for ML models.

What we're looking for

• BS in Computer Science or related Engineering with 10+ years of experience or an MS degree with 6+ years of experience.
• Extensive experience in Python or cloud computing and AWS.
• Understanding of quantization techniques such as PTQ and AWQ for accelerating LLM inference.
• Solid understanding of Transformer models and the challenges of serving large transformer-based models.
• Solid understanding of ML training, especially reinforcement learning algorithms like GRPO and DAPO.
• Experience in systems engineering fundamentals including caching, CUDA, autoscaling, high throughput, and low latency.
• Experience in monitoring and observability tools to monitor model input/output and feature stats.
• Experience with recommendation systems, containers (Docker, Kubernetes, ECS), Ray, vLLM, and RL libraries (preferred).