Staff Machine Learning Engineer, Censys

• Bachelor’s degree in Computer Science, Data Science, Engineering, or other technical discipline (or equivalent professional experience)
• 5+ years of experience in docker, Kubernetes, and Helm
• Strong proficiency in Python and machine learning libraries like PyTorch, Transformers, and Timm
• Strong proficiency in MLOps tooling like Metaflow, MLflow, Argo Workflows, Torchrun, and Ray
• Experience with streaming data processing frameworks like Kafka and Spark
• Experience with applying model optimization techniques such as quantization, pruning, and distillation to improve performance
• Experience working with cloud platforms like AWS, GCP, and Azure

Desirable

• Censys is looking for a Staff Machine Learning Engineer to join our team and help us derive valuable insights from internet security datasets
• This role is responsible for building and maintaining a robust internal Machine Learning Operations (MLOps) platform capable of handling petabyte-scale data and delivering high-throughput, low-latency predictions
• You will engineer and optimize this platform to support a diverse range of machine learning applications, including computer vision, natural language processing, and reinforcement learning, ensuring seamless deployment and scalability
• We’re a highly collaborative, creative team dedicated to empowering customers with innovative solutions to improve their security posture
• You’ll work closely with data scientists, engineers, and product teams to build scalable systems that enhance how customers interact with and understand our data
• You will design machine learning systems that not only process and analyze data but also bring critical insights to the surface in a way that empowers decision-making
• Deploy and maintain containerized workloads to support machine learning development, deployment, and post-deployment monitoring
• Utilize tools like Helm and Kustomize to accelerate the deployment of machine learning models and data pipelines
• Apply various optimization techniques such as compilation, quantization-aware training (QAT), and pruning to improve the latency and throughput of models
• Utilize open-source software like Metaflow, Prefect, Temporal, and Argo Workflows to facilitate data science development
• Build and optimize machine learning models to analyze security data, extract actionable insights, and identify trends, anomalies, and other relevant security signals
• Develop and maintain systems for drift detection and model monitoring to ensure continuous improvement and accuracy of insights
• Collaborate with cross-functional teams to design data pipelines that can efficiently process petabytes of raw internet security data