Fleet Scheduler Data Scientist

AST SpaceMobile•Lanham, MD

12d•Onsite

About The Position

AST SpaceMobile is building the first and only global cellular broadband network in space to operate directly with standard, unmodified mobile devices based on our extensive IP and patent portfolio and designed for both commercial and government applications. Our engineers and space scientists are on a mission to eliminate the connectivity gaps faced by today’s five billion mobile subscribers and finally bring broadband to the billions who remain unconnected. We are seeking a Fleet Scheduler Data Scientist to define, build, and operationalize algorithms that optimize and control this network. This role will lead the design and deployment of AI and ML powered large-scale combinatorial optimization algorithms that drive predictive planning and reactive real-time control loops. This role requires a deep understanding of state-of-the-art techniques in combinatorial optimization of heterogeneous graphs for system scheduling, and must develop an understanding of a host of underlying systems and subsystems to efficiently incorporate their features into both predictive and reactive action planning loops. This person will work in coordination with software engineers to implement the algorithms and models that are developed, and will also work closely with aerospace, network, and operations engineering teams to translate physics- and constraint-driven system behavior into scalable prediction, optimization, and control architectures.

Requirements

BS/MS/PhD in Computer Science, Electrical Engineering, Applied Math, Physics, Aerospace, or a related field, or equivalent experience.
5-7+ years delivering AI, ML, and other optimization systems in production environments.
Demonstrated experience solving large-scale combinatorial optimization problems (e.g., scheduling, resource allocation, logistics, network capacity).
Strong background in traditional optimization methods, including mixed-integer programming, constraint programming, or related techniques.
Experience modeling and solving large-scale problems represented as graphs or heterogeneous graph structures.
Strong Python and ML/optimization tooling (e.g., PyTorch).
Experience building data pipelines and deploying models into operational systems.
Strong cross-functional collaboration skills, partnering effectively with software, aerospace, network, and operations engineering teams.
Strong communication skills, with the ability to translate complex physics- and constraint-driven system behavior into clear technical requirements.
Strong analytical and problem-solving skills, with the ability to identify and drive new AI/ML opportunities across design, manufacturing, and operations.
Meticulous attention to detail in model validation, production readiness, and drift monitoring.

Nice To Haves

Experience with reinforcement learning, multi-agent systems, or hybrid ML + optimization.
Familiarity with satellite operations and RF communications.
Experience building real-time or safety-critical decision systems.
Prior technical leadership of AI/ML or optimization teams.

Responsibilities

Own the architecture of AI and ML powered satellite and ground scheduling algorithms.
Design and deploy large-scale combinatorial optimization algorithms, with approaches ranging from mixed-integer programming and heuristics to reinforcement-learning-based solvers operating on heterogeneous graphs.
Incorporate state-of-the-art techniques in combinatorial optimization, learning-augmented planning, graph-based scheduling, and computational efficiency.
Create a multi-tiered scheduling system, from high-level control of major satellite activities through mid-level control of system configurations and settings.
Develop physics-informed ML models that accurately predict satellite behaviors, utilizing simulated spacecraft model data and satellite telemetry data for fine-tuning.
Ensure all models and optimizers are production-grade: data pipelines, training, validation, deployment, monitoring, and drift management.
Partner with satellite engineers to develop simulated training data, define key system physics, and ensure models reflect true system behavior and constraints.
Coordinate with satellite engineers to incorporate fleet modeling into design analysis and trades.
Identify and drive new opportunities for AI and ML across design, manufacturing, and in-orbit operations.