Machine Learning Engineer II

Role Overview

Join Abnormal AI’s Message Detection – Attack Detection team to build high-recall detection systems that identify evolving email/communication attacks. The team combines behavioral signals, generalizable and auto-trained models, and category-specific detectors to improve efficacy against new and unseen threats.

What You Will Do

• Design and implement systems that combine rules, ML models, and feature engineering with product/business inputs to power an email detection product (with senior engineer guidance).
• Understand differentiating signals between safe emails and attacks, and how the model stack supports detection.
• Identify and propose new feature groups and/or ML model approaches to improve detection efficacy.
• Work with infrastructure/systems engineers to productionize signals feeding the detection system.
• Write production-quality code with emphasis on testability, readability, edge cases, and error handling.
• Train models on well-defined datasets to improve performance on specialized attacks.
• Monitor and improve false negative (FN) rates and overall efficacy using feature engineering, rules, and ML modeling.
• Analyze FN/FP datasets to identify capability gaps and recommend short-term improvements via features/rules.
• Contribute across the stack as needed, including data pipelines and presenting results in internal tools.

Must Have

• 3+ years experience designing, building, and deploying machine learning applications in domains such as:
  • text understanding / entity recognition / NLP
  • computer vision
  • recommendation systems
  • search
• 1+ years of experience writing/stabl** (posting text is truncated).