Machine Learning Engineer

Client Content Safety team builds machine learning systems that help keep client safe, welcoming, and age-appropriate for hundreds of millions of users around the world. We develop and operate production ML systems that help identify potentially harmful, sensitive, or policy-relevant content across multiple formats, including audio, text, images, and video. Our work supports teams across client in making responsible content decisions, improving user experiences, and meeting evolving safety and regulatory expectations. This is high-impact applied ML work at large scale. You’ll join a collaborative team working on challenging problems in content understanding, multimodal classification, model evaluation, and responsible AI. As client continues to expand the ways creators and users interact with content, our systems play an important role in helping those experiences launch and grow safely. What you’ll do • Design, build, and improve production ML systems that help detect harmful, sensitive, or policy-relevant content across multiple modalities. • Own model development end to end, from training data design and experimentation through evaluation, deployment, and production monitoring. • Build and improve multimodal detection approaches using modern machine learning techniques, including vision, text, audio, and large language model-based systems where appropriate. • Develop robust evaluation frameworks, including standardized datasets, labeling guidance, precision and recall analysis, threshold calibration, and monitoring for model performance over time. • Partner with cross-functional teams, including Trust & Safety, Policy, Legal, Public Affairs, Product, and Engineering, to translate content safety needs into scalable technical solutions. • Drive experimentation to improve model quality, reliability, fairness, and safety impact in real-world production settings. • Contribute to the team’s technical strategy, including model architecture, signal design, evaluation methodology, and ML infrastructure choices. • Mentor other engineers and help raise the bar for ML craft, evaluation rigor, and responsible AI practices. Who you are • You have strong experience building, deploying, and operating machine learning systems in production environments. • Minimum 5-6 years of experience. • You are comfortable working with modern ML frameworks such as PyTorch and have experience with vision, text, audio, or multimodal models. • You have a strong understanding of ML evaluation, including dataset design, labeling methodology, threshold calibration, and precision/recall trade-offs. • You know how to design ML systems that balance model quality, reliability, latency, scalability, and product impact. • You care deeply about responsible ML and think carefully about failure modes, bias, edge cases, and the impact of model decisions on users and creators. • You enjoy working across disciplines and can collaborate effectively with policy, legal, product, and trust & safety stakeholders. • You have experience leading technical projects and influencing technical direction within a team or product area. • Experience with content moderation, content safety, policy-driven ML, or responsible AI systems is a plus, but not required.