Staff Machine Learning Engineer

Integral Ad Science is a global technology and data company that builds verification, optimization, and analytics solutions for the advertising industry. The Staff Machine Learning Engineer will provide technical leadership for AI/ML services, architect large-scale machine learning infrastructure, translate business needs into scalable systems, and mentor engineering and data science teams.

Responsibilities:

• Technical Leadership & Vision: Drive the architectural vision and system design for our core AI/ML-based services
• Act as the technical lead for complex, multi-quarter initiatives from inception to global deployment
• Architect at Scale: Design and build large-scale deep learning infrastructure and platforms for distributed model training, ensuring low-latency and high-availability at enterprise scale
• Cross-Functional Influence: Partner with Product Management, Core Engineering, and executive stakeholders to align ML capabilities with business strategy
• Translate abstract product requirements into concrete technical designs
• Act as a Multiplier: Mentor and guide senior and mid-level data scientists and engineers
• Establish standard methodologies, define code quality expectations, and lead architecture design reviews
• Infrastructure Mastery: Work with large-scale AI training infra components (accelerators, network fabrics, CUDA, NCCL, RDMA) and big data ecosystems (Databricks, Spark, Kubernetes, Kafka, Prometheus)
• End-to-End Ownership: Design, develop, and support robust CI/CD pipelines for AI/ML services, ensuring models transition smoothly from research into reliable production endpoints

Requirements:

• PhD/Master's degree in a technical field such as computer science, mathematics, statistics, or equivalent years of experience
• 7+ years of machine learning experience in industry, with a proven track record of operating at a Lead, Principal, or Staff level
• Deep expertise in designing complex machine learning systems, ranking infrastructures, and data pipelines that handle massive throughput
• Extensive, production-hardened experience working with frameworks such as PyT orch, JAX, or T ensorFlow
• Strong, production-grade programming skills in Python, Go, Java, or C++, along with a deep understanding of software design principles and algorithms
• Proven ability to thrive in ambiguity, take large unstructured problems, and independently drive them to successful technical resolutions
• A strong history of collaborating with, elevating, and mentoring engineering talent