Staff GNC Engineer (State Estimation)

Turning Space into a Transportation Layer for Earth Who We Are: Inversion builds advanced reentry systems to deliver next-generation capabilities from space. Our mission is to make Earth radically more accessible by turning Low-Earth Orbit into an on-demand logistics domain. We see space not as a destination, but as a platform — one that unlocks unprecedented speed and global reach. Our spacecraft are designed to deliver payloads anywhere on Earth in under an hour, operating through extreme reentry conditions and landing with high precision. These systems open the door to new ways of testing, delivering, and operating at hypersonic speeds. Inherently dual-use, our technology is built to meet urgent national security needs while laying the groundwork for future commercial applications. Backed by leading investors including Y Combinator, Spark Capital, and Lockheed Martin Ventures, and working with partners such as the U.S. Space Force and NASA, Inversion is pushing the boundaries of what's possible in space-based defense and logistics. What You'll Do: Inversion's vehicles do not fly through an empty sky. As the Staff GNC Engineer (State Estimation), you will give our vehicles an accurate, continuously updated picture of the world around them — estimating and predicting the state and behavior of friendly and non-cooperative systems external to the vehicle. This problem shares DNA with the prediction stacks that let autonomous cars anticipate the paths of surrounding vehicles and pedestrians, and you will draw on both classical estimation theory and modern learned methods to solve it in a far more demanding flight regime. In this role, you will: Develop state estimation and tracking algorithms for aerospace systems external to the vehicle, spanning cooperative platforms and non-cooperative objects observed only through onboard sensor measurements Model and predict the behavior of external systems, including maneuvering objects with uncertain intent Train, validate, and deploy neural network models for trajectory and behavior prediction, and integrate them alongside classical filtering approaches Develop probabilistic representations of external-object state and intent that downstream guidance and planning functions can consume Build the metrics, tooling, and datasets needed to quantify estimation and prediction error and drive systematic improvement Integrate estimation and prediction algorithms into 3-DOF and 6-DOF simulation and carry them through real-time flight software Work closely with the guidance, sensors, and simulation teams to close vehicle-level performance Required Qualifications: Bachelor's degree in Aerospace Engineering, Electrical Engineering, Robotics, a related field, or equivalent experience Typically, 9+ years of applicable experience developing and testing estimation, tracking, or GNC algorithms and systems Experience with behavior or trajectory prediction for autonomous vehicles, robotics, or similar multi-agent domains, including probabilistic prediction of agent intent Experience with modern deep learning frameworks (e.g., PyTorch, JAX) and the infrastructure to train models at scale Experience estimating and tracking the state of dynamic objects from noisy, intermittent, or limited sensor data Experience training and implementing neural networks for prediction, tracking, or related applications Solid grasp of classical mechanics, dynamics, and rigid body motion Proficiency in programming languages such as Python, MATLAB, or C++ for simulation and analysis Demonstrated excellent verbal and written communication skills Capable of working in a dynamic, fast-paced startup environment This position is on-site at Inversion HQ in Playa Vista, CA. Must have the ability to obtain and maintain a U.S. government Secret/Top Secret security clearance. Desired Qualifications: Master's or PhD in Aerospace Engineering, Electrical Engineering, Robotics, a related field, or equivalent experience Experience with vehicle…