Senior Security Data Engineer, Censys

• We’re looking for someone with an ownership mindset, is curious, can provide mentorship, and has a collaborative spirit for our mission
• 5+ years of experience in research or software engineering with a security data focus
• Experience analyzing large datasets and turning ambiguous data into usable information that solves security problems
• Strong programming skills in Go/Python (or similar), plus experience using SQL (or similar) tools to analyze large datasets
• Strong communication skills and the ability to work effectively with researchers, ML engineers, software engineers, and other security experts
• Comfort working with security-relevant data and applying technical judgment to questions about Internet-exposed hosts, services, and infrastructure

Desirable

• We’re looking to hire a Senior Security Data Engineer to analyze large-scale Internet data and help build the features, labels, and datasets that power our security-focused Artificial Intelligence (AI) / Machine Learning (ML) models
• In this role, you will help turn raw Internet telemetry into enriched, model-ready data that improves how we classify and understand hosts, services, infrastructure, and more across the Internet
• Your core focus areas will support such systems so they have the labelled security data required to classify entities as benign, suspicious, or malicious
• This will add further AI powered solutions to support our Security Operations Center (SOC) initiatives
• This role is intentionally distinct from a SOC analyst, threat hunter and rapid response type work; instead, it focuses on the context side of the problem: transforming labels, expert input, and telemetry into scalable model inputs and security-relevant data products
• Analyze Censys telemetry and derived datasets to identify signals that improve AI/ML model training for classification that affects security outcomes
• Build and improve training and evaluation datasets using Internet telemetry, manually curated labels, and analyst-reviewed data
• Drive feature discovery, feature selection, and labelling strategies for models that classify entities as benign, suspicious, or malicious
• Work on multi-layer labeling and classification problems, where categories such as device type, router, honeypot, or edge service may need to be identified before risk classification
• Partner with Research / Detection teams to translate security domain expertise into actionable workflows
• Collaborate with ML engineers and software engineers to ensure features, labels, and model inputs are practical to productionize
• Contribute to feedback loops and evaluation frameworks that improve precision, recall, confidence, and coverage over time
• Build tooling to support the efforts listed above