SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 135,000 - 170,000 / annual
Lodestar Space is developing autonomous defense capabilities for high-value space assets in orbit. The company's flagship product, MITHRIL, is an AI-enabled autonomy software suite that enables spacecraft to autonomously detect, characterize, and neutralize orbital threats.
As a Software Engineer: State Estimation & Prediction, you will lead development of core state estimation and prediction models powering MITHRIL. You'll research and develop algorithms that fuse probabilistic estimation with machine learning to track targets, predict trajectories, and assess intent in real-time autonomous spacecraft operations.
Key responsibilities include:
- Design and implement state estimation and prediction architecture for autonomous spacecraft
- Research and implement novel estimation and prediction algorithms from literature review through deployment
- Develop classical and neural estimators to track multiple dynamic targets in real time
- Create neural models to forecast future trajectories and behavioral patterns
- Implement intent inference models to identify actions and rank threat levels
- Integrate models into mission simulation environments and autonomy decision systems
- Collaborate with perception and autonomy teams to ensure prediction fidelity and scalability
Required qualifications: Bachelor's or Master's in Computer Science, Aerospace, Robotics, Applied Mathematics, or equivalent; 2+ years industry experience in state estimation, aerospace, or robotics; strong C++ and Python proficiency; demonstrated experience with probabilistic state estimation (Kalman filters, particle filters, Bayesian inference); experience with deep learning frameworks (PyTorch, TensorFlow) for sequence modeling; familiarity with trajectory modeling or orbital mechanics.
Preferred: track record implementing state estimation in real-time/safety-critical systems; experience deploying seq2seq models for forecasting; sensor fusion techniques; GPU acceleration (CUDA, TensorRT); Linux, Git, CI/CD; Docker/Kubernetes; real-time systems and performance optimization; networking for distributed autonomy.
ITAR compliance required: must be U.S. citizen, lawful permanent resident, or eligible for State Department authorization.
