Machine Learning Engineer I - Message Security Products

Overview

Machine Learning Engineer I for Abnormal AI’s Misdirected Email Detection (MED) team, building end-to-end ML solutions to detect and block misdirected outbound emails.

What you'll do

• Own the full ML lifecycle for misdirected email detection, from data wrangling to monitoring.
• Develop and refine features, run experiments, and improve detection reliability.
• Conduct offline metrics, online A/B testing, and post-launch monitoring with threshold setting and error analysis.
• Deploy and maintain production models and detection systems, including drift/performance monitoring and rollback guardrails.
• Collaborate with Product Manager, Tech Lead, and engineering stakeholders to match technical deliverables to roadmap milestones.
• Participate in a shared on-call rotation focused on detection efficacy and real-time scoring systems.
• Investigate efficacy-related alerts and high-visibility false positives/false negatives.

What you'll need

• BS degree in Computer Science, Machine Learning, Artificial Intelligence, Information Systems, or a related quantitative field.
• 1+ years building and operating applied ML features in production systems.
• Experience contributing to end-to-end ML systems (data wrangling, feature engineering, training, evaluation, deployment, and monitoring).
• Ability to implement and reason about algorithms and use numerical computing for signal combination/averaging.
• Experience interrogating production data, identifying behavioral/trend shifts, and launching targeted experiments.
• Understanding of online vs offline pipelines and labeling/workflow concepts for safe, scalable deployments.
• Experience running offline metrics, online A/B tests, threshold setting, and monitoring drift/performance.
• Strong written and asynchronous communication skills.
• Ability to work independently and across distributed, cross-functional teams.

Nice to have

• Experience with Python, Go, AWS, Spark, or Databricks.
• Experience in email security/DLP or misdirected email prevention domains.
• Experience writing detectors/rules to complement ML models.
• Experience operationalizing research into customer-facing systems with scalability and detection accuracy focus.
• Experience contributing to a small team delivering a component from scratch.