I obtained my PhD in computational structural mechanics from CSMLab at Cambridge University, where my research focused on isogeometric analysis using subdivision techniques and its applications in large structural deformations, nematic elastomers and dynamic fractures. During my postdoctoral research, I have extended my interests from deterministic problems to stochastic problems, for example, Bayesian experimental design, sparse quadratures for stochastic pdes and Bayesian model validation. My recent work on fast optimal experimental design has been focused on using Laplace approximation, sparse quadratures and multi level methods to accelerate the computation of the information theoretic utility function, the so-called expected information gain, for experiments associated with large scale PDEs. The new methodology can significantly improve the computational efficiency, which is the major factor hindering the application of Bayesian optimal experimental design in engineering. I have applied the cost-efficient method to the design of current patterns in impedance tomography, the sensor placements for seismic source inversion, the design of optimal initial temperature and fuel concentration in combustion experiments, and the detectability of transverse cracks in laminated composites.