Sagar Doshi Email & Phone Number

Cambridge, Massachusetts, United States

• Built high voltage probing setup and made electrode mask layout (gdsfactory) to pole thin-film lithium niobate (TFLN)• Simulated (eigenmode solver) waveguide cross sections for photonics inverse design of nonlinear frequency converters• Optimized second harmonic generation (SHG) microscope and waveguide probing setup to characterize TFLN devices• Demonstrated state-of-the-art quasi phase matching gratings in SiN-on-TFLN and etched TFLN waveguides, measured SHG• Simulated… Show more • Built high voltage probing setup and made electrode mask layout (gdsfactory) to pole thin-film lithium niobate (TFLN)• Simulated (eigenmode solver) waveguide cross sections for photonics inverse design of nonlinear frequency converters• Optimized second harmonic generation (SHG) microscope and waveguide probing setup to characterize TFLN devices• Demonstrated state-of-the-art quasi phase matching gratings in SiN-on-TFLN and etched TFLN waveguides, measured SHG• Simulated device electrostatics (COMSOL Multiphysics®) to engineer periodic ferroelectric domain (FE) inversion Show less

Cambridge, Massachusetts, United States

• Developed two-photon microscope software (GUI, back-end, control) anddesigned mechanical components (objective changer, optical layout)• Managed clinical trial NCT02926729 Real-Time Assessment Of BreastCancer Lumpectomy Specimen Margins With Nonlinear Microscopy (16-145)• Collected simulated nerve sparing data for NLM evaluation or radicalprostatectomies and designed interventional study under R01-CA249151-01• Collected, processed, and organized two-photon imaging data… Show more • Developed two-photon microscope software (GUI, back-end, control) anddesigned mechanical components (objective changer, optical layout)• Managed clinical trial NCT02926729 Real-Time Assessment Of BreastCancer Lumpectomy Specimen Margins With Nonlinear Microscopy (16-145)• Collected simulated nerve sparing data for NLM evaluation or radicalprostatectomies and designed interventional study under R01-CA249151-01• Collected, processed, and organized two-photon imaging data and H&E slidesfor neuro, gyn, and hematopoietic pathology studies with clinical collaborators Show less

Stanford, California, United States

• Pipelined real-world patient data, simulation data into GPU-based imagereconstruction algorithm. Configured reconstruction to optimize runtime.• Automated parameter generation and modeled PETCT detector in SimSET• Automated PET/CT data simulation, image reconstruction, andtraining/testing Pix2Pix CNN (PyTorch)

Ucla

• Implemented, trained, and tested capsule-based CNN to virtually stainhistological samples imaged using FLI microscopy, benchmarked with U-Net• Researched image processing techniques for bright field and autofluorescencelifetime microscopy, tested CNN for low-photon conditions

Manhattan, New York

• Implemented Clustering & Regression analysis on trade data(250,000+ trades/month) with Python scripts (TensorFlow, Pandas, NumPy)• Developed automated HTML email report generator using Flask and Jinja2• Developed GUI to display FENICS pricing data using ReactJS, used AJAX (fetchAPI) to refresh pricing data in real time

Doctor Of Philosophy - Phd, Electrical And Electronics Engineering

Master Of Science - Ms, Electrical And Electronics Engineering

MS Thesis: Nonlinear Microscopy for Materials Analysis and Clinical Pathology

Bachelor Of Science - Bs, Electrical And Electronics Engineering

Bachelor Of Science - Bs, Applied Mathematics