Research Scientist
Constellation Space
Research Scientist
Constellation Space
AI-driven OS for satellite network management
We are looking for a PhD-level Research Scientist to build the models behind ConstellationOS. You will develop forecasting and decision models for satellite link conditions and ground-network operations, and prove them against real data and physics-based baselines. Your work shows up in customer pilots and government evaluations.
Develop and train machine learning models that forecast link conditions and support operational decisions.
Own data acquisition and preprocessing for telemetry, weather, atmospheric, and RF datasets.
Benchmark learned models against deterministic physics models and report results honestly, including failures.
Do simulation research for GNC and RF systems.
Work with academic groups and publish or present results
Turn research into code the platform team can ship.
Ph.D. (or near completion) in Machine Learning, Applied Mathematics, Physics, Aerospace, Electrical Engineering, or a related field.
Publications or equivalent research output in forecasting, deep learning, reinforcement learning, or scientific ML.
Strong Python and PyTorch or JAX. Able to write research code others can run.
Experience with time-series, signal, or physical-system data is a plus.
U.S. person required (ITAR).
AI-driven OS for satellite network management
Constellation Space builds ConstellationOS, an AI-powered operating system for large satellite networks. It acts as an autonomous infrastructure layer that ingests telemetry from satellites, ground stations, and weather systems to predict link failures with over 90% accuracy, issuing a five-minute warning and autonomously rerouting traffic in under two seconds to prevent data loss without human intervention. The platform targets satellite operators and seeks to replace reactive NOC playbooks by enabling proactive, automated network management at mega-constellation scale. The team comes from SpaceX, Blue Origin, and NASA, lending domain experience, and the company aims to reduce industry losses from network failures (estimated at $2.5 billion annually) by improving resilience and efficiency. Financially, the company progressed from Y Combinator's Winter 2026 batch with a $500K accelerator round to a $6.5M seed in March 2026 led by SGH Capital, to fund platform development and expansion.
Explore more opportunities at Constellation Space