Senior Machine Learning Engineer

About the Role

Abnormal Security is seeking a Senior Machine Learning Engineer to become a part of the Message Detection Decisioning team. At Abnormal, we strive to protect our customers from malicious adversaries who persistently enhance their strategies and techniques to outsmart and undercut conventional security procedures. This is what makes our original behavioral-based strategy so...Abnormal. Abnormal has continually been ranked among the cybersecurity startups, and our behavior AI system has helped us secure various awards, resulting in being entrusted to shield over 8% of the Fortune 1000 (and growing).

In an environment where a single successful attack can lead to financial damages amounting to millions of dollars, the Message Detection Decisioning team takes the central role in developing a high-precision Detection Engine capable of operating on hundreds of millions of messages at millisecond latencies. All emails processed by Abnormal pass through the workflow managed by the Detection Decisioning team, which applies hundreds of signals and detectors to a message, depending on the context of the message and the user. The system then calculates the final overall decision for the system and subsequently records attribution to steer various offline and online metrics such as offline precision, online precision, online False Negative Rate, and so on.

The team is tackling a multilayered detection problem, which involves modeling communication patterns to establish baselines across the enterprise. By incorporating these patterns as robust signals and combining them with contextual information, they create highly precision systems. The team builds discriminatory signals at various levels including message level (e.g., presence of specific phrases), sender level (e.g., sender's frequency), and recipient level (e.g., likelihood of receiving a safe message), which serves as the bedrock to build highly accurate heuristic and model-based detectors. To maintain a consistently high-precision detection system, the team also introduces innovation in software systems and processes that can be rapidly adapted to solve short-term trends and generalize well in the long run.

This position also presents the opportunity to have a considerable impact on the team's overall charter, direction, and growth. The Senior Machine Learning Engineer will be involved in understanding the customer’s most pressing problems in the domain of false positives and building a related technical roadmap to keep our detection decisioning system operating at an extremely high precision.

What you will do 

• Design and put into operation systems that integrate rules, models, feature engineering, and business and product inputs into an email detection product.
• Identify and recommend new feature groups or ML model approaches that could substantially improve detection efficacy for a product. Work with infrastructure & systems engineers to transform signals for ingestion by the detection system.
• Understand features that distinguish between safe emails and email attacks, and how our detector stack allows us to identify them.
• Become an expert in the main detection pipelines and decision data flow to drive debugging in systematic degradations induced by poor detectors.
• Write code considering testability, readability, edge cases, and errors.
• Train models on defined datasets to enhance model efficacy on specialized attacks.
• Proactively monitor and improve false positive rates and efficacy rates for our message detection product attack categories, through feature engineering, rules, and ML modeling.
• Analyze false negative and false positive datasets to categorize capability gaps and to recommend short-term feature and rule ideas to enhance our detection efficacy.
• Contribute to other areas of the stack by building and debugging data pipelines, or presenting results back to customers in our tools when required.
• Lead the team's medium and long-term roadmap and steer planning and execution strategy for the pod.
• Coach and mentor junior engineers to elevate their code quality and ML effectiveness through quality code reviews and design reviews.
• Participate in creating a world-class detection engine across all layers - data quality, feature engineering, model development, experimentation, and operation.

Must Haves 

• A proven track record of turning business requirements into scalable, maintainable systems with a bias towards simpler, iterative systems.
• 4+ years of experience with production ML systems - understands the pillars of a modern ML stack and the design, maintenance, and tuning processes of ML models.
• An ability to systematically debug both data and system issues within ML/heuristics models.
• Proficiency in Python and machine learning libraries like numpy and scikit-learn.
• Experience with data analytics and proficiency with SQL+pandas+spark framework to both create data and metric generation pipelines, and to address critical questions about system efficacy or counterfactual treatments.
• A history of independently managing the entire lifecycle of projects or features including engineering design, development, and deployment.
• Experience collaborating with other stakeholders - has worked with cross-functional teams to steer projects to completion.
• An academic background in machine learning (Bachelor's degree in Computer Science or related fields).

Nice to Have 

• An MS degree in Computer Science, Electrical Engineering, or another related engineering field
• Experience with big data or statistics
• Familiarity with the cybersecurity industry

This position is not: 

• A role focused on optimizing existing machine learning models
• A research-oriented role that's distanced from the product or customer
• A combination of statistics/data science and ML role

#LI-ML1