Machine Learning Engineer

Northstar is a seed-stage cybersecurity startup building an AI-native security platform. Security teams struggle to understand and defend the complex environments that they are responsible for. We’re fixing that. Northstar connects data from the systems an organization uses to build a model of its environment, then uses AI to help security teams understand risk and take action. We’re looking for a talented machine learning engineer to help uncover insights, surface hidden risks, and improve security decision-making using the data we bring together. You’ll join a small team of outstanding engineers and play a central role in our ML efforts: helping identify and shape ML opportunities, finding useful signals in imperfect data, building and improving models, and demonstrating that they make the product better. You’ll work closely with product and engineering to bring models into production, and evaluate how they perform with real security teams. What we’re looking for Hands-on experience developing, evaluating, and improving machine learning models for real-world problems Strong machine learning and statistics fundamentals, with the judgment to choose appropriate approaches - from traditional ML and statistical methods to deep learning and LLMs - based on the problem and available data. A track record of owning modeling decisions: what you built, why you chose the approach, how you evaluated it, and what you changed when it fell short Comfort working with complex, imperfect data, limited ground truth, and defining useful targets, features, and measures of success Experience designing evaluations that account for false positives, missed risks, and the consequences of incorrect results Ability to independently implement and experiment with models, and work with engineers to deploy, evaluate, and improve them in production. Bias for ownership, learning, and making progress when the problem, data, or success criteria aren’t fully defined Interest in an early-stage role with hands-on responsibility and influence over technical direction Your current title might be applied scientist, data scientist, ML engineer, or AI engineer. Your experience might include risk scoring, anomaly detection, ranking, recommendation, forecasting, and other applied ML problems. These are examples, not a checklist. Cybersecurity experience is welcome but not required. We care more about your work and judgment than a particular degree, employer, or number of years of experience. This is a hands-on applied ML role. LLMs may be part of the solution, but we're looking for someone with experience building and evaluating models beyond prompt engineering and LLM application development. Follow us