Principal Machine Learning Engineer

Equinix is the world’s digital infrastructure company, enabling connectivity and digital innovation across global environments. The Principal Machine Learning Engineer will design, build, deploy, and scale production-grade machine learning and generative AI systems across multi-cloud platforms. The role combines applied machine learning, software engineering, and MLOps, including developing end-to-end pipelines, LLM-powered systems, and solutions for NLP and computer vision.

Responsibilities:

• Design, develop, and deploy machine learning and Large Language Model (LLM)–based solutions for production use cases
• Collaborate with Generative AI Center of Excellence leaders and business stakeholders to evaluate buy vs. build decisions for generative AI applications
• Develop end-to-end ML pipelines, covering data ingestion, feature engineering, model training, evaluation, deployment, and monitoring
• Architect and implement LLM-powered systems that integrate agents and services across multiple cloud platforms into a unified solution
• Optimize ML workflows for performance, scalability, reliability, and cost efficiency in cloud environments (GCP, Azure, AWS)
• Implement and maintain MLOps best practices, including CI/CD, model versioning, experiment tracking, and automated retraining
• Work extensively with deep learning frameworks such as PyTorch and TensorFlow
• Containerize ML services and deploy them using Docker, Kubernetes, App Engine, or virtual machines
• Apply strong knowledge of NLP fundamentals, including transformers, attention mechanisms, embeddings, and text preprocessing
• Deploy and manage models in production, conduct A/B testing, and measure performance improvements using statistical methods
• Develop features, run experiments, analyze results, and translate insights into actionable improvements
• Build and deploy classical ML models (regression, classification, clustering), NLP applications (sentiment analysis, summarization, Q&A, chatbots, information retrieval), and computer vision solutions (image classification, object detection, segmentation using models such as YOLOv7, DDRNet, RFTM with datasets like COCO and Cityscapes)