We are looking for a Senior or Staff Software Engineer to improve behavior prediction for autonomous driving. In this role, you will work on how our autonomy system predicts the behavior, intent, and future motion of other road users, and how those predictions are used by the planning system to make safe, comfortable, and effective decisions. This role is ideal for someone with deep experience in autonomous vehicle prediction, trajectory prediction, behavior modeling, or actor intent understanding. You will work closely with Planning and ML engineers to improve prediction quality, model integration, evaluation, and downstream planning behavior. This is not only an ML integration role. Through this high-impact role, you will improve prediction using the most effective approach for the problem, whether that means improving heuristic prediction logic, collaborating on learned prediction models, refining prediction-planning interfaces, or changing how Planning consumes prediction outputs. What You'll Do • Own prediction improvements within the Planning team. Advance prediction models and collaborate closely with ML teams on learned prediction models, model outputs, uncertainty, and failure modes. • Improve prediction-planning interfaces, including confidence, uncertainty, intent representation, and fallback behavior. • Collaborate on projects that improve planning behavior in scenarios where actor intent is uncertain or changing. • Define metrics and evaluation frameworks for prediction quality as it relates to planning behavior. Analyze prediction-related planning failures using logs, simulation, metrics, and scenario-based evaluation. • Lead technical projects and mentor junior engineers. What We're Looking For • Strong experience with behavior prediction, trajectory prediction, actor intent modeling, or related autonomous vehicle prediction problems. • Strong software engineering skills with proficiency in C++. Python proficiency is a plus. • Experience working with planning systems and understanding of how prediction impacts planning, safety, comfort, and system behavior. • Ability to analyze complex autonomy failures using logs, metrics, and simulation. • Strong collaboration skills across Perception, Planning, ML, Simulation, and Systems teams. • Strong system-level judgment around prediction uncertainty, safety, and deployment and ability to reason across heuristic, learned, and hybrid prediction approaches. The pay range listed below reflects the base salary in our SF/Silicon Valley location, across several internal levels. Actual starting pay will be based on job-related factors including: work location, experience, relevant training, education, skill level and performance during interview. Total compensation at Kodiak includes base pay, equity, bonus and a competitive benefits package California Pay Range $240,000-$265,000 USD