Lead Machine Learning Engineering

What you'll do

• Build and maintain scalable data pipelines for the full lifecycle of LLM development and production deployment.
• Architect and manage human-in-the-loop labeling workflows to ensure high-quality training data.
• Develop strategies for synthetic data generation, filtering, and validation to improve dataset diversity.
• Use LLMs and machine learning techniques to automate data generation, scoring, and evaluation processes.
• Establish systems to measure and mitigate dataset failure modes such as bias, contamination, and distribution shifts.
• Design experiments to link specific dataset compositions to measurable improvements in model performance.
• Provide technical direction on infrastructure, compute, and storage decisions for machine learning projects.
• Mentor team members and conduct design reviews to ensure engineering excellence across the project.

What we're looking for

• Bachelor's degree in a STEM field with 8+ years of experience, Master's with 6+ years, or PhD with 3+ years of relevant experience.
• 3+ years of hands-on experience building, curating, and scaling datasets for machine learning training and evaluation.
• 5+ years of professional programming experience using Python, C++, or Go in a production or research environment.
• 5+ years of experience using frameworks like PyTorch or TensorFlow to develop, train, evaluate, and deploy ML models.
• Expertise in curating and managing datasets for the LLM lifecycle, including synthetic data generation and post-training workflows (preferred).
• Proficiency in designing human-in-the-loop labeling systems and mitigating dataset failure modes like bias and contamination (preferred).
• Demonstrated success using LLMs for data generation and model-assisted labeling to improve performance (preferred).
• Strong technical foundation in distributed data processing frameworks such as Spark, Ray, or Beam (preferred).