Physical AI Engineer - SW

📋 Description

• Build/train/validate ML models that approximate physics with uncertainty.
• Generate synthetic datasets for training and stress-testing models.
• Integrate models with CFD/FEA and optimization tools for fast answers.
• Integrate foundation models into engineering workflows with human oversight.
• Create reliable, traceable ML outputs for engineers to trust.
• Ship research to production in a typed Python monorepo with reproducibility.

🎯 Requirements

• Strong Python skills and ML pipelines in real, version-controlled codebases.
• Hands-on ML: train, evaluate, and debug models for production-ready implementations.
• Knowledge of physics-informed ML, neural operators, or surrogate modeling.
• Experience generating or using synthetic data to train systems.
• Judgment on foundation models; know when to trust or verify.
• Evidence-first mindset; explicit data/verification for outputs.
• BSc/MSc or equivalent in quantitative/engineering field.
• Solid Git and modern software practices.
• Excellent cross-disciplinary communication and collaboration.
• Genuine aviation interest and learning systems for real-world use.

🎁 Benefits

• Aerospace background; CFD/FEA or MDAO familiarity.
• Differentiable optimization or enforcing physical constraints in learned models.
• Exposure to safety-critical or regulated environments.
• Sim-to-real techniques and reconciling with hardware/flight data.
• Lab instrumentation experience (oscilloscopes, analyzers, HIL/SIL rigs).
• Fluency with scientific Python and ML stacks (PyTorch/JAX, async services, queues, time-series DBs).
• Understanding model-scaling principles and trade-offs.