Staff Security ML Researcher

We need to output HTML fragment with sections: About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule. Only include sections that have real specific content. We must not repeat job title. Must not include Job ID etc. Must remove boilerplate, navigation, legal, EEO, scam warning, duplicated lines, application instructions. Must preserve all substantive info. We need to parse the posting. Let's extract sections: - Intro: "Onwards Together! Illumio is the leader..." etc. That's about company. Might be considered "About the role"? Actually "About The Role" heading appears later. The intro before that is generic company description; could be part of "About the role"? Usually "About the role" includes description of role. The posting has "About The Role" then description. So we should include that as

About the role

with paragraphs. - Then "Your Impact" with subheadings: Machine Learning & Security Intelligence, Threat Research & Risk Modeling, Data Science & Analytics Engineering, Product Collaboration & Innovation, Research & Thought Leadership. Those could be considered responsibilities. So we can put under

Responsibilities

with

• each bullet? The content is paragraphs under each subheading. We could treat each subheading as a responsibility bullet. But we need to use
  • for any list of duties/requirements/benefits. So we can create a list of responsibilities items. Each bullet could be a paragraph? Usually
  • can contain

    . We'll output

  • ...

  . Ensure we preserve all substantive info. - Then "Your Toolkit" with Required Qualifications and Preferred Qualifications and Bonus Points. Those map to Requirements, Qualifications, Skills maybe. We need to decide sections: Requirements (maybe required qualifications), Qualifications (maybe preferred?), Skills (maybe toolkit?). The instruction: Use headings for section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule). So we need to map content accordingly. We have Required Qualifications list: bullet points. That's likely Requirements. Preferred Qualifications list: maybe Qualifications. Bonus Points: maybe Skills? Or could be additional qualifications. But we need to avoid thin sections. We could combine Required and Preferred into Requirements and Qualifications sections. Skills could be extracted from toolkit: "Strong Python programming skills, including experience with data science and machine learning frameworks such as Pandas, NumPy, Scikit-learn, TensorFlow, or PyTorch." etc. That's part of Required Qualifications already. So maybe we don't need separate Skills section if content overlaps. But we could still include Skills if there is distinct content: The Required Qualifications already list skills. The Preferred Qualifications list also includes experience etc. Bonus Points list includes background, publications, certifications. Those could be considered Skills or Additional Qualifications. However instruction: Prefer a few substantive sections over many thin ones. So maybe we combine Required and Preferred into Requirements (or Qualifications) and skip Skills if not distinct. But we need to include sections only when posting has real specific content. The posting does have Required Qualifications, Preferred Qualifications, Bonus Points. So we could have: - Requirements: Required Qualifications (list) - Qualifications: Preferred Qualifications (list) - Skills: Bonus Points (list) maybe. But we need to ensure we don't create sections with filler like "commensurate with experience". Not the case. Also we have location: "Location: 4 on-site days a week in Sunnyvale, CA Headquarters." That's schedule? Actually schedule could be "4 on-site days a week". That's schedule info. Also location maybe not needed as heading. Could be placed in a short

  under About the role? But we have to preserve all substantive info. We could include location in a paragraph under About the role or maybe under a separate section? There's no heading for location. Could add a

  after About the role before Responsibilities. That's okay. Also there is "Our Team's Vision" etc. That's maybe part of About the role? Could be included in About the role paragraph. We must remove boilerplate: The sections like "Our Commitment", "All official job offers...", "Please be aware...", "Upon acceptance...", "If you receive any communication...", "Your security is important...", "For roles in San Francisco and Los Angeles: ..." These are legal/EEO etc. Must be removed. Also remove duplicated lines, navigation, cookie, etc. Now we need to output only HTML fragment. Let's craft:

  About the role

  ... (combined intro + Our Team's Vision + About The Role description)

  Then location maybe as a separate paragraph:

  Location: 4 on-site days a week in Sunnyvale, CA Headquarters.

  Then Responsibilities: list each bullet from Your Impact subheadings. We need to preserve all substantive info from each subheading. Let's extract each subheading content: Machine Learning & Security Intelligence: - Design, develop, and deploy machine learning models that identify anomalous behavior, detect threats, and predict security risk across complex enterprise environments. - Apply statistical techniques and predictive modeling to evaluate breach likelihood, attack exposure, and segmentation effectiveness. - Build behavioral profiling, anomaly detection, clustering, classification, and forecasting models using large-scale security telemetry. - Evaluate model performance, reduce false positives, and continuously improve detection quality through experimentation and data-driven analysis. Threat Research & Risk Modeling: - Analyze large-scale security datasets to identify attacker behaviors, emerging threats, and patterns aligned to frameworks such as MITRE ATT&CK. - Develop threat scoring and risk assessment models that help customers prioritize remediation and security investments. - Leverage graph-based analysis techniques to model attack paths, quantify lateral movement risk, and recommend mitigation strategies. - Work closely with threat researchers to transform security intelligence into measurable detection and prediction capabilities. Data Science & Analytics Engineering: - Design and maintain scalable data pipelines used for model training, validation, and analytics. - Investigate new features, signals, and telemetry sources to improve detection accuracy and risk predictions. - Partner with engineers to productionize models and analytics systems while ensuring scalability, reliability, and observability. - Conduct experiments and hypothesis-driven analysis to validate new detection approaches and security insights. Product Collaboration & Innovation: - Collaborate with product managers, designers, and engineers to embed ML-driven security insights into customer-facing experiences. - Influence product strategy through data-backed recommendations and threat intelligence findings. - Help define security analytics frameworks, data requirements, and detection methodologies for future product capabilities. - Serve as a subject matter expert on the use of machine learning in cybersecurity applications. Research & Thought Leadership: - Explore emerging techniques in machine learning, graph analytics, and AI-driven threat detection. - Investigate applications of graph neural networks, probabilistic modeling, and advanced analytics for cyber defense. - Contribute to patents, technical publications, conference presentations, and thought leadership initiatives. - Stay current on advancements in cybersecurity, machine learning, and adversary tradecraft to continuously evolve Illumio's detection capabilities. Now Required Qualifications list: - 5+ years of experience in security analytics, threat research, detection engineering or data science, machine learning - Strong Python programming skills, including experience with data science and machine learning frameworks such as Pandas, NumPy, Scikit-learn, TensorFlow, or PyTorch. - Experience designing, training, validating, and deploying machine learning or statistical models in production environments. - Proven ability to work with large-scale telemetry datasets from security, networking, cloud, or infrastructure environments. - Strong understanding of model evaluation, feature engineering, experimentation, and handling imbalanced datasets. - Experience applying analytics or machine learning techniques to cybersecurity problems such as anomaly detection, behavioral analysis, threat hunting, or risk assessment. - Familiarity with security frameworks such as MITRE ATT&CK and common security telemetry sources. - Strong SQL and data querying skills. - Excellent communication skills with experience translating technical findings into actionable business and product recommendations. Preferred Qualifications: - 7-10+ years of experience spanning cybersecurity, machine learning, data science, or threat research. - Experience building graph-based security analytics using Neo4j, graph databases, or graph algorithms. - Knowledge of graph machine learning techniques, including graph neural networks or link prediction methods. - Experience productionizing ML models in cloud environments using AWS, Kubernetes, or similar platforms. - Experience developing risk-scoring systems, recommendation engines, or predictive analytics solutions. - MS or PhD in Computer Science, Data Science, Machine Learning, Cybersecurity, Statistics, or a related field. Bonus Points: - Background at a cybersecurity company focused on cloud security, network security, endpoint security, or threat intelligence. - Experience integrating external threat intelligence feeds and enrichment data into ML workflows. - Publications, patents, open-source contributions, or conference presentations related to security or applied machine learning. - Industry certifications such as CISSP, GIAC, or advanced machine learning credentials. Now we need to map to sections. We could do:

  About the role

  ... (intro + Our Team's Vision + About The Role description)

  Location: 4 on-site days a week in Sunnyvale, CA Headquarters.

  Responsibilities

  • Machine Learning & Security Intelligence: Design, develop, and deploy machine learning models that identify anomalous behavior, detect threats, and predict security risk across complex enterprise environments. Apply statistical techniques and predictive modeling to evaluate breach likelihood, attack exposure, and segmentation effectiveness. Build behavioral profiling, anomaly detection, clustering, classification, and forecasting models using large-scale security telemetry. Evaluate model performance, reduce false positives, and continuously improve detection quality through experimentation and data-driven analysis.

  • Threat Research & Risk Modeling: Analyze large-scale security datasets to identify attacker behaviors, emerging threats, and patterns aligned to frameworks such as MITRE ATT&CK. Develop threat scoring and risk assessment models that help customers prioritize remediation and security investments. Leverage graph-based analysis techniques to model attack paths, quantify lateral movement risk, and recommend mitigation strategies. Work closely with threat researchers to transform security intelligence into measurable detection and prediction capabilities.

  • Data Science & Analytics Engineering: Design and maintain scalable data pipelines used for model training, validation, and analytics. Investigate new features, signals, and telemetry sources to improve detection accuracy and risk predictions. Partner with engineers to productionize models and analytics systems while ensuring scalability, reliability, and observability. Conduct experiments and hypothesis-driven analysis to validate new detection approaches and security insights.

  • Product Collaboration & Innovation: Collaborate with product managers, designers, and engineers to embed ML-driven security insights into customer-facing experiences. Influence product strategy through data-backed recommendations and threat intelligence findings. Help define security analytics frameworks, data requirements, and detection methodologies for future product capabilities. Serve as a subject matter expert on the use of machine learning in cybersecurity applications.

  • Research & Thought Leadership: Explore emerging techniques in machine learning, graph analytics, and AI-driven threat detection. Investigate applications of graph neural networks, probabilistic modeling, and advanced analytics for cyber defense. Contribute to patents, technical publications, conference presentations, and thought leadership initiatives. Stay current on advancements in cybersecurity, machine learning, and adversary tradecraft to continuously evolve Illumio's detection capabilities.

  Requirements

  • 5+ years of experience in security analytics, threat research, detection engineering or data science, machine learning.

  • Strong Python programming skills, including experience with data science and machine learning frameworks such as Pandas, NumPy, Scikit-learn, TensorFlow, or PyTorch.

  • Experience designing, training, validating, and deploying machine learning or statistical models in production environments.

  • Proven ability to work with large-scale telemetry datasets from security, networking, cloud, or infrastructure environments.

  • Strong understanding of model evaluation, feature engineering