Build a Physics-Informed Neural Operator (PINO) to Accelerate a Python Fluid Dynamics Solver

I am looking for an experienced Machine Learning Engineer or Researcher specializing in Scientific Machine Learning (SciML) to help accelerate a 1D fluid dynamics propagation solver. I already have a fully functioning, high-fidelity numerical solver written in Python that is validated against reference data. The goal of this project is to build an AI-native hybrid pipeline that uses a Physics-Informed Neural Operator (PINO) or Fourier Neural Operator (FNO) to speed up the simulation. Deliverables: #Develop the AI Operator: Build a 1D neural operator in PyTorch that can instantly predict the evolution of fluid/pressure waveforms across different conditions. #Embed Physics Constraints: Incorporate the governing differential equations (Burgers Equation physics) into the model's loss function so the AI outputs physically realistic results. #Create a Hybrid Pipeline: Integrate the AI model with my existing Python solver, creating a smart trigger that switches between the fast AI prediction and the high-fidelity numerical solver when sharp gradients or shockwaves form. Requirements: 1. Strong experience with PyTorch or JAX. 2. Proven experience building physics-informed machine learning models (PINO, FNO, or DeepONet). 3. Familiarity with training models on physical/synthetic data generated from numerical code.