Perception & State Estimation Engineer - DoorDash Air

About the Team DoorDash Labs is an independent team within DoorDash. We explore robotics and automation to transform last mile logistics in the long term. If you have a passion for applying robotics solutions in a service used by millions of people, then we want to talk to you! About the Role We're hiring a State Estimation & Perception Engineer in our Autonomy Software team. In this role, you'll own and develop the perception and state estimation for our package pickup and delivery system. You'll influence system design decisions, including sensor selection, sensing architecture, and observability analysis, at the earliest stage when choices still matter. You're excited about this opportunity because you will… • Play an integral role on a small and focused team • Own tuning and implementation of the state estimation stack on the drone (PX4 EKF modifications, fault tolerance, multi-sensor fusion with VIO and laser odometry) • Own state estimation and perception of the delivery mechanism, including dynamics modeling, observation model design, and sensor selection • Work closely with controls and planning to close the loop on automated pickup and delivery • Work closely with the hardware team to define sensing principles and sensor placement • Design and execute flight test campaigns to validate estimator performance • Identify and resolve failure modes in estimation (degenerate geometries, sensor dropout, initialization) We're excited about you because… • You have deep experience with state estimation and filtering (EKF, UKF, MSCKF, particle filters) • You have experience with optimization-based methods (factor graphs, GTSAM, Ceres, SymForce) • You have experience with visual-inertial odometry (VIO) and/or laser odometry (LIO) • You're proficient in modern computer vision (feature extraction, tracking, depth estimation) • You have strong C++ and/or Rust skills, with Python for prototyping and analysis • You have experience deploying estimation on real hardware with real sensor noise • You can derive an EKF from scratch, including process model, measurement model, observability analysis, and consistency checks • You have deep linear algebra knowledge (matrix decompositions, sparsity exploitation, numerical conditioning) • You've closed a perception-to-control loop on real hardware under degraded conditions • You're comfortable operating at the intersection of theory and embedded reality • You're driven by a passion for delivering outstanding customer experiences • You have a proactive approach to development, often crafting new tools • MS/PhD degree in CS, EE, ME, Robotics, or related technical field Nice to have experience includes • Experience with PX4 / ArduPilot EKF internals • Experience with multi-body state estimation (tethered or suspended payloads) • Familiarity with SLAM in GPS-denied environments • Publications or open-source contributions in state estimation or VIO • You've debugged estimator failures in the field, not just in simulation Applications for this position are accepted on an ongoing basis Compensation The successful candidate’s starting pay will fall within the pay range listed below and is determined based on job-related factors including, but not limited to, skills, experience, qualifications, work location, and market conditions. Base salary is localized according to an employee’s work location. Ranges are market-dependent and may be modified in the future. In addition to base salary, the compensation for this role includes opportunities for equity grants. Talk to your recruiter for more information. DoorDash cares about you and your overall well-being. That’s why we offer a comprehensive benefits package to all regular employees, which includes a 401(k) plan with employe