Senior Data Scientist ML Classification & Content Safety

Key Responsibilities:

• Define the evaluation methodology for content classification at Securly and ensure that every model release meets the established standard before shipping. - Lead the multiclass refactor of Securly's content classification models, transitioning binary models to handle multi-label, multi-class content categories such as Adult Content, Violence, Self-Harm, Social Media, and others. - Build and maintain labeled evaluation datasets with robust annotation workflows, addressing class imbalance and label noise systematically, and documenting dataset curation decisions in a versioned data card. - Connect offline evaluation to production monitoring to identify classification drift and error patterns proactively before they impact customers. - Investigate and resolve misclassification errors, including false positives (over-blocking) and false negatives (under-blocking), and produce written root cause analyses. - Mentor the existing AI team on evaluation methodology, model development practices, and data science communication rigor. - Collaborate with engineering to integrate model outputs into the production filtering stack while considering appropriate latency and reliability constraints. - Research and prototype improvements such as feature representations, model architectures, active learning for label efficiency, and domain adaptation for emerging content categories. Qualifications Required:

• Machine learning experience in multi-label/multi-class classification, model evaluation methodology, handling class imbalance, and text/URL data feature engineering with a minimum of 5 years in applied ML roles. - Proficiency in Python (ML stack) for production-quality code, including scikit-learn, PyTorch or TensorFlow, pandas, and numpy, with experience in using notebooks for exploration and production-grade pipelines for delivery. - Strong understanding of text/NLP feature engineering, ML evaluation rigor, data engineering for ML, technical communication, and stakeholder influence. - Experience with large-scale classification in production, active learning/annotation workflows, cloud ML infrastructure, web content/URL classification domain, and familiarity with K-12/CIPA compliance. - Ability to communicate precision/recall tradeoffs effectively to different stakeholders and produce executive-level summaries of classification quality for leadership. Key Responsibilities: