Senior Machine Learning Engineer, Shield
Job Description
At Box, the team behind Box Shield helps deliver enterprise security machine learning that can detect threats, suspicious behavior, and malicious content without slowing down the work your organization depends on. This role focuses on building and deploying ML capabilities that support production detection at scale, with the added expectation of participating in an on-call rotation to help triage and respond to issues as they arise.
What you’ll do
- Design, train, and deploy machine learning models for ransomware detection, suspicious session identification, and user behavior analytics, including anomaly detection.
- Own end-to-end ML pipelines for high-volume security event streams using Apache Spark, GCP Dataflow, GCP Dataproc, BigQuery, and Vertex AI.
- Create and maintain feature stores that support both real-time and batch anomaly detection systems.
- Deploy, monitor, and iterate on production models, serving enterprise customers at scale.
- Collaborate cross-functionally with Platform, Application Engineering, and Product teams to translate security requirements into ML solutions.
- Join the on-call rotation and be available while on-call to help respond to and triage issues that arise.
What you bring
- 5+ years of experience in applied machine learning.
- Experience leading design and implementation of scalable ML systems across build, deploy, and support.
- GCP experience (Vertex AI, BigQuery, Dataflow) or equivalent experience with AWS SageMaker or Azure ML.
- Strong communication skills and the ability to explain complex ML concepts to non-technical stakeholders.
- An ownership mindset focused on high-quality technical execution and effective collaboration.
- Bachelor’s degree in Computer Science (or equivalent practical experience).
- Strong ML fundamentals covering model training and evaluation (including weights, loss functions, precision, recall, F1, and calibration), feature engineering, embeddings, and real-world data issues such as class imbalance, label noise, and model drift. Experience limited to generative AI tools, agents, or APIs does not meet this bar.
- Deep understanding of MLOps, monitoring, and updating models that serve real traffic.
- Clear, inclusive communication and a focus on collaboration, mentorship, and continuous improvement.
Helpful experience
- Security/threat or fraud detection experience, especially with sequential data and behavioral modeling (anomaly detection, time-series forecasting, LSTM, Transformers, or similar).
- Experience with streaming or real-time ML systems.
- Experience with FedRAMP or other compliance-constrained environments.
- Familiarity with Java stack integrations.
- Publications or contributions in ML security.
Tech you’ll work with
Languages: Python, Go, Java
ML/Data: Apache Spark, Vertex AI, BigQuery, TensorFlow/PyTorch
Infrastructure: GCP, Kubernetes
Domains: User Behavior Analytics, Time-Series Anomaly Detection, AI Security, Content classification, Ransomware Detection
Location: Redwood City, CA (onsite)
Salary: USD 211,000 - 263,500 per yearly
Office expectation: Boxers are expected to work from their assigned office a minimum of 3 days per week.