<p>Hydroforming is used in sheet metal manufacturing for creating complex shapes and lightweight designs. Still, understanding its process parameters relies on widespread experimentation and time consuming simulations. This study represents a Physics Informed Neural Network (PINN) surrogate that aims to predict two crucial hydroforming outputs, dome height, and blank diameter. Trained on 21 data points, the proposed approach incorporates a simplified physics inspired constraint intended to regularize learning into its training objective, rather than a full description of the hydroforming process. Dome height predictions show agreement with experimental measurements, giving an R² of 1.000 and a root mean square error (RMSE) of 0.0147&#xa0;mm. Blank diameter prediction, which is more inspiring due to complex material flow remaining stable (R² = 0.9415, RMSE = 0.2901&#xa0;mm). The training losses show rapid initial improvement, followed by brief modifications as the network balances the objectives. While the present formulation adopts simplified assumptions and is not intended to replace detailed finite element simulations, the approach shows potential as a surrogate model for preliminary analysis, trend prediction, and early-stage process exploration in hydroforming. Although, the current model has simplified assumptions, the framework can be further refined as additional data and physics are incorporated. Future work will focus on expanding the experimental dataset and investigating adaptive loss weighting, transfer learning across multiple different geometries, and integrating multi fidelity simulation data. This preliminary study demonstrates that combining data with the first principles of physics through PINNs is an early, yet important step.</p>