-Used Simulink and Embedded Coder to develop, build, and test auto-generated embedded code for the company's 2nd generation cooling wristband within 3 months of joining.-Achieved a 25% improvement in lumped device heat transfer model temperature forecasting using grey box models from the MATLAB system identification toolbox and a 150x decrease in simulation time.-Verified the accuracy of the thermoelectric cooler's manufacturer performance specifications using a testbed that I co-developed with my mechanical engineer colleague.-Developed a prototype nonlinear model predictive control strategy using thermal sensation models from academic literature and demonstrated a 10% reduction in device heating for a desired stimulus level.-Used device session data to create a novel metric that proved fundamental to device thermal performance benchmarking.-Leveraged an unused battery current sensor signal to develop a power system electrical resistance recursive least squares estimation routine to detect faults and monitor performance degradation.

-Achieved up to a 20% reduction of peak power on new customer pilots using MATLAB's Statistics and Machine Learning toolbox to calculate peak load prediction models and evaluate peak demand and energy consumption savings. -Saved hours of manual effort for each customer equipment change using Python's pandas toolbox to automate the maintenance of our equipment database.-Weathered the loss of 3 research team members as the Principal Investigator on a Department of Energy SBIR Phase I Grant and delivered the final report on time.-Hired and trained a contractor to support the short-staffed grant research team.-Used MATLAB to develop the company's first generation economic predictive control algorithm for charging electric vehicles. In a pilot study, the algorithm achieved a 37% reduction in peak demand relative to the facility's baseline operation strategy.

-Developed a total cost of ownership optimization strategy that leveraged the European trailer system's new variable speed condenser fan. Using the modeling team's MATLAB refrigeration system and Simulink trailer thermal models to develop a fan speed lookup table implementable on an embedded processor.-Contributed results to team effort to persuade the European trailer platform team to adopt the methodology.

-Developed embedded flight control software requirements in Rational DOORS for a continuous built-in test (CBIT) to isolate failed sensors and actuators from the redundant surface control system.-Used the Simulink system model to perform validation of the CBIT performance against customer fault recovery requirements.-Validated all CBIT software and system level requirements in the hydraulic systems lab using dSPACE Control Desk automated test procedures.

I focused on application of extremum seeking control algorithms to energy consumption minimization in vapor compression air conditioning and refrigeration systems.-Contributed to and helped maintain Thermosys library of ~20 vapor compression system component MATLAB/Simulink models.-First author on 2 American Control Conference papers (2016 and 2017)-see publications section for details-Wrote a Master's thesis that focused on:1. A comparison between 3 different extremum seeking algorithms applied to vapor compression systems.2. Using self-optimizing control and classical extremum seeking control to improve the performance of extremum seeking.

Focused on predictive modeling and optimization of a building chiller plant's power consumption using a data driven (machine learning) approach.

I have worked on improving the company's vapor compression system dynamic modeling software, Thermosys.

I investigated carbon fiber laminate buckling using ABAQUS finite element package

I helped facilitate tests of quasi-statically indented 16 ply carbon fiber laminate, aluminum honeycomb sandwich panels and performed shadow moire analysis on test videos.