Cybersecurity Data Scientist - Automation & AI (Remote)

at Merck in Boise, Idaho, United States

Job Description

Position Summary:

We are hiring a Cybersecurity Data Scientist to join the Cybersecurity Automation & AI team. This is a builder role inside a cybersecurity engineering organization-not a research lab, not an analytics reporting team. Your job is to develop real models that drive containment, response, and automation at scale using telemetry from platforms like Microsoft Defender XDR , Sentinel, Wiz, and ServiceNow.

Job Description:

This role is about impact. You will work on systems that directly affect enterprise risk posture, automating decisions around isolation, escalation, and prioritization. The models you build must work reliably inside operational pipelines, integrate with engineering workflows, and be explainable under pressure.

We are building adaptive, system-aware automation-models that reason over behavior and drive real-time action. That means everything you create must be contextual, actionable, and robust enough to operate in live production environments.

If youre used to exploratory notebooks with no consequences or havent worked with security data before, this role will be hard. You must understand the stakes, the complexity, and how automation changes enterprise behavior. Were not looking for theoretical modelers-we need someone who can think in systems and ship.

Key Responsibilities:

+ Build and maintain ML scoring logic in Databricks using telemetry from security platforms

+ Engineer behavior models, anomaly detectors, or confidence scoring systems that directly support automation

+ Collaborate with security engineers to embed models into workflows across Defender, Sentinel, ServiceNow, and other platforms

+ Think critically about automation safety, control boundaries, and unintended consequences

+ Validate and tune models based on stakeholder input and real-world telemetry

+ Document model logic, assumptions, edge cases, and operational safety mechanisms

+ Work with platform and automation engineers to integrate outputs cleanly into orchestration layers

Required Qualifications:

+ 1-2 years of experience in applied data science, machine learning, or automation-focused analytics

+ Proficient in Python and libraries like pandas, scikit-learn, and XGBoost

+ Hands-on experience with Databricks or similar environments for pipeline development

+ Strong working knowledge of telemetry, logs, time series, or event-based data

+ Foundational cybersecurity knowledge-if you dont understand