• Physics informed Machine Learning: Developed and implemented a physics informed machine learning model for particle based systems with TensorFlow and Keras. Successfully incorporated neural network into the physics based model to enhance the efficiency of simulations by ~28% in smoothed particle hydrodynamics (SPH).• Data pre-processing: Automated labelling and processing of simulation data to create structured dataset of 130k data points.• Dimensionality reduction of dataset: Used ANN with particles within the kernel instead of all particles, thus significantly reducing the dimensionality of the system (from 1000 particles to 30 particles).• Test-driven Software development: Implemented, validated and incorporated 2 new models into the codebase, enhancing the software’s capabilities, expanding the application range and enabling several new use cases.• Optimization: Modeled ring laser beam, achieving optimization for additive manufacturing, which notably reduced the experimental-simulation mismatch by ~5% .• Effective Communication: Presented research findings to international research team of ~20 researchers at seminars conveying complex ideas to diverse audiences.

• Data Generation and Processing: Implemented a Monte-Carlo algorithm with a molecular dynamics simulator to generate ~50,000 labelled data points. Pre-processed and transformed the atomic co-ordinates to images with Keras.• Computer Vision: Employed Convolutional Neural Networks using Keras to classify core-shell and homogeneous nano-particles. Achieved F1 score of 0.952 through hyperparameter optimization.

• Data generation: data collection from experiments using SEM, XRD, and mechanical testing on 2 3D printed metal samples (IN718 & IN625).• Data analytics: investigated and analyzed data with Python and NumPy to derive valuable insights.• Communication: of research data and conclusions to technical and non-technical staff at the company.