Staff AI/ML Engineer

Role Overview

As a Staff AI/ML Engineer at Dragos, you will design and implement production machine learning systems for the Dragos platform. You will work closely with Data Engineers and product teams to deliver AI/ML capabilities that strengthen threat detection and security analysis for ICS (Industrial Control Systems) and xOT (Extended Operational Technology) cybersecurity.

Responsibilities

• Design and implement production-grade ML systems that work in both cloud and resource-constrained on-premises environments.
• Build and optimize ML model architectures for ICS/OT cybersecurity use cases including:
  • Threat detection
  • Asset classification
  • Behavioral analysis
  • Anomaly detection
  • Natural language processing (NLP)
• Develop robust data pipelines and ML workflows supporting real-time and batch processing.
• Collaborate with OT detection experts to translate research prototypes into scalable production systems.
• Partner with Data Engineers to establish data contracts and implement ML observability (monitoring, versioning, deployment best practices).
• Improve ML infrastructure, including:
  • automated testing frameworks
  • CI/CD pipelines
  • deployment strategies for containerized environments (Kubernetes, Docker)
• Evaluate and adapt state-of-the-art ML research and open-source models for domain-specific cybersecurity applications.
• Troubleshoot and optimize production ML performance, focusing on latency, accuracy, and resource utilization.

Qualifications

• 6+ years of engineering experience, including 4+ years focused on production ML implementations.
• Strong software engineering foundation with expertise in Python and SQL, plus experience with at least one additional language: Go, Rust, Java, or JVM-family languages.
• Experience building and deploying ML systems using modern frameworks such as scikit-learn, PyTorch, TensorFlow, and/or HuggingFace.
• Demonstrated experience with ML solutions such as classification, time series analysis, anomaly detection, or NLP that drive measurable impact.
• Experience with MLOps practices: model versioning, monitoring, pipeline orchestration, and deployment in high-reliability environments.
• Familiarity with data engineering concepts: pipelines, stream processing, message queuing, and medium-to-large datasets.
• Knowledge of containerized deployment and cloud-native architectures.
• Strong communication skills and ability to collaborate with diverse stakeholders.

Nice-to-haves

• Experience with LLMs, retrieval-augmented generation (RAG), or advanced NLP techniques.