Navigation Engineer; State Estimation​/Orbit Determination

Location

Mountain View or San Diego

Full time

On-site

Engineering

• $130K – $250K

Logos Space is building a proliferated Low Earth Orbit (pLEO) constellation to deliver resilient, high-performance connectivity to enterprise and government users worldwide. The Navigation team builds the software and infrastructure for spacecraft timing, astrodynamics, orbit determination, and the associated navigation workflows that support both flight and ground systems.

The Navigation team at Logos Space is building the software that determines where our spacecraft are now and into the future, and how that information flows through flight and ground systems to support resilient constellation operations. This role focuses on the design, implementation, and validation of the underlying state-estimation capabilities required to operate our proliferated LEO constellation.

In this role, you will develop core estimation software used across onboard real-time estimation, ground-based orbit determination, analysis, and validation tooling. Depending on level, you will contribute to or independently own well-scoped portions of the estimation stack, such as process and measurement modeling, filter design and validation, covariance analysis, estimator integrity, statistical validation, replay, or regression infrastructure.

We are looking for engineers who can work well in a fast-paced environment, turn incomplete requirements into practical software, and collaborate closely with adjacent teams.

• Develop and maintain in-house C++ libraries for state estimation, orbit determination, and related navigation workflows.

• Build Python tools and/or bindings for analysis, testing, automation, and rapid debugging.

• Implement and validate process and measurement models, filtering and smoothing logic, covariance propagation, initialization, gating, and related numerical utilities.

• Integrate data from GNSS, timing sources, maneuvers, telemetry, and external tracking inputs.

• Build simulation, replay, Monte Carlo, and regression tooling to verify estimator performance and catch numerical or integration regressions.

• Analyze estimator consistency, convergence, observability, initialization behavior, and off-nominal performance.

• Work closely with flight software, GNSS, astrodynamics, timing, and mission operations engineers to define and implement interfaces.

• Support integrated test campaigns, commissioning, and flight-data investigations.

• Bachelor’s degree in Aerospace Engineering, Computer Science, Electrical Engineering, Physics, Applied Mathematics, or a related STEM field, or equivalent practical experience.

• 2+ years of relevant professional experience building navigation, estimation, orbit determination, GNSS, robotics localization, or similar numerical software. Strong graduate research or project experience may substitute for some professional experience.

• Proficiency in modern C++ for production software development.

• Proficiency in Python for analysis, testing, automation, and data investigation.

• Working knowledge of Kalman filtering, least-squares estimation, covariance propagation, and statistical performance analysis.

• Working knowledge of linear algebra, numerical methods, and debugging numerical or floating-point issues.

• Familiarity with orbital mechanics, reference frames, and time systems, or similar navigation fundamentals in another domain.

• Experience with Linux-based development environments, Git, and automated testing.

• Strong written and verbal communication skills.

• 4+ years of relevant experience and the ability to independently own technical work streams.

• Experience with extended Kalman filters, batch least-squares orbit determination, smoothing, factor graphs, or related estimation methods.

• Experience with LEO spacecraft navigation, GNSS observation modeling, spacecraft orbit determination workflows, or timing-sensitive estimation problems.

• Experience with real-time or resource-constrained software.

• Experience building reproducible analysis…