Senior Applied AI/ML Engineer – Digital Twin & Space Missions

About the role

Space Cargo Unlimited is building AI‑driven digital twins to enable research and manufacturing in microgravity. As a Senior Applied AI/ML Engineer you will lead the design, development and deployment of advanced machine‑learning and generative‑AI models that power mission preparation, in‑orbit operations and scientific data analysis.

Key responsibilities

• Design, train, fine‑tune and ship ML, deep‑learning and generative‑AI models (PyTorch, TensorFlow, JAX) for vision, time‑series, multimodal and scientific data.
• Architect physics‑informed and hybrid AI‑simulation systems (PINNs, neural surrogates, neural radiance fields, reduced‑order models) that combine first‑principles physics with data‑driven approaches.
• Own end‑to‑end components of the digital‑twin stack, from data ingestion and simulation coupling to model serving, monitoring and continuous improvement.
• Translate state‑of‑the‑art research into robust, production‑ready prototypes and internal tools, deciding what to build versus defer.
• Process experimental, telemetry and simulation data to improve model accuracy, calibration and trustworthiness for mission‑critical use.
• Collaborate with systems engineers and flight‑operations teams to ensure AI solutions meet verification, safety and operational requirements.
• Contribute to technical road‑mapping, architectural decisions and hiring; mentor junior engineers and review their work.

Required profile

• MSc, engineering degree or PhD in Applied Mathematics, Computer Science, Physics, Aerospace or a related discipline.
• 8 + years of hands‑on experience building, training and deploying AI/ML/DL systems in industrial or applied‑research environments.
• Proven ability to work autonomously on complex, mission‑critical projects and to guide junior team members.

Required skills

• Python programming.
• Deep‑learning frameworks: PyTorch, TensorFlow, JAX.
• Physics‑informed neural networks (PINNs) and hybrid AI‑simulation techniques.
• Neural surrogates, neural radiance fields, reduced‑order models.
• Model deployment, serving, monitoring and data‑pipeline engineering.