Maintained backend Golang pipeline which processed daily WiFi and camera data from Raspberry Pi devices and populated SQL tables with calculated insights. Also updated pipeline with new features in response to customer requests.

Analyzed runtime efficiency of backend pipeline using Golang pprof profiler, and improved efficiency by parallelizing long loops and rewriting algorithms to use much less memory.

Proposed an algorithm using camera data instead of WiFi data to calculate average visitor dwelling time, which vastly improved the accuracy of results.

Explored and tested Prometheus monitoring solution for use with Raspberry Pi devices as well as applications running in Google Cloud VM.

Evaluated the accuracy of algorithms using WiFi data by modifying input parameters as well as internal algorithm logic, and then comparing the resulting trends over a month with camera data results.

Used Paraview (an open-source, multi-platform application designed to visualize data sets) to generate scripts for use with VPIC Catalyst In-situ Adapter. Installed and tested VPIC (vector particle in-cell code).

Created kernels in OpenCL for functions to execute on GPU computing devices to parallelize a large code written in C/C++ for dynamical simulations of dusty astrophysical systems.

Used Python with yt tool to plot 2D/3D images of the density, magnetic field and velocity distributions.

Developed Perl scripts to process genomic data. Installed, configured and tested bioinformatics tools.

Built websites with HTML, CSS and JavaScript for Oralgen web-based database system and genome analysis tools.

Held office hours for individual student discussion.

Graded assignments.