Machine Learning Engineer II

About the role

Abnormal AI is looking for a Machine Learning Engineer to join the Message Detection - Attack Detection team. At Abnormal, we protect our customers against nefarious adversaries who are constantly evolving their techniques and tactics to outwit and undermine traditional approaches to Security. Our novel behavioral-based approach has earned us recognition as one of the top cybersecurity startups, with a behavioral AI system trusted to protect more than 25% of the Fortune 500.

In a landscape where a single successful attack can lead to financial losses of millions of dollars, the Attack Detection team plays the central role of building an extremely high recall Detection Engine that can operate on hundreds of millions of messages at milliseconds latency. The team’s mission is to provide world-class detector efficacy to tackle the changing attack landscape using a combination of generalizable and auto-trained models, as well as specific detectors for high-value attack categories.

This team solves a multi-layered detection problem, modeling communication patterns to establish enterprise-wide baselines, incorporating these patterns as robust signals, and combining them with contextual information to create precise systems. The team builds discriminative signals at various levels—message (e.g., presence of particular phrases), sender (e.g., frequency of sender), and recipient (e.g., likelihood of receiving a safe message)—and combines them to train highly accurate model-based and heuristic detectors. To adapt to new unseen attacks, the team builds automated model retraining pipelines, including data analytics, generation, modeling, production evaluation, and deployment stages.

This role offers the opportunity to significantly impact the team’s charter, direction, and roadmap. The Machine Learning Engineer will analyze false negatives—current and future attacks that could disrupt customer workflows—define the technical roadmap to address pressing customer problems, and operate the detection system at an extremely high recall.

Responsibilities

• Design and implement systems that combine rules, models, feature engineering, and business/product inputs into an email detection product, with senior engineer guidance.
• Understand features that distinguish safe emails from email attacks and how the model stack enables detection.
• Identify and recommend new feature groups or ML model approaches to improve detection efficacy.
• Work with infrastructure and systems engineers to productionize signals for the detection system.
• Write code with testability, readability, edge cases, and errors in mind.
• Train models on well-defined datasets to improve efficacy on specialized attacks.
• Actively monitor and improve false negative (FN) and efficacy rates through feature engineering, rules, and ML modeling.
• Analyze FN and false positive (FP) datasets to categorize capability gaps and recommend short-term feature and rule improvements.
• Contribute to other areas of the stack, such as building and debugging data pipelines or presenting results to customers.

Requirements

• 3+ years of experience designing, building, and deploying machine learning applications in domains such as text understanding, entity recognition, NLP, computer vision, recommendation systems, or search.
• 1+ years of experience writing stable, production-level pipelines for model training and evaluation, ensuring reproducible models and metrics.
• Experience with data analytics and using SQL, pandas, and Spark to build data/metric generation pipelines and answer critical questions about system efficacy or counterfactual treatments.
• Ability to thoroughly understand business requirements and design the simplest yet generalizable ML model/system to achieve goals.
• Systematic approach to debugging data and system issues within ML/heuristic models.
• Fluency in Python and machine learning toolkits like NumPy, scikit-learn, PyTorch, and TensorFlow.
• Strong software engineering skills, including quickly finding answers in the codebase and writing structured, readable, well-tested, and efficient code.
• BS degree in Computer Science, Applied Sciences, Information Systems, or a related engineering field.

Nice to Have

• MS degree in Computer Science, Electrical Engineering, or a related engineering field.
• Experience with big data, statistics, and machine learning.
• Experience with algorithms and optimization.

This position is not a role focused on optimizing existing machine learning models, a research-oriented role removed from the product or customer, or a statistics/data science-focused ML role.

Pay

Base salary range: $160,700—$231,000 USD. Actual compensation will be determined based on skills, experience, qualifications, and geographic location. In addition to base salary, this role may be eligible for bonus or incentive compensation, equity, and a comprehensive benefits package.

Abnormal AI uses AI-assisted tools to help our recruiting team prepare for candidate interviews by analyzing resume content and role requirements to suggest interview questions and identify areas for exploration. These tools do not make hiring decisions or automatically screen candidates—every decision is made by a person.