My engineering expertise ranges across a wide variety of fields and applications:
- Design Optimization, Design for Manufacturing and Assembly, Structural and Thermal Modeling for Finite Element Analysis (ANSYS), CAD Modeling (SolidWorks), Composite Materials (Fiberglass), Rapid Manufacturing Technologies (including 3D printing), and Lean Manufacturing
- Printed Circuit Board Design, Electronics Prototyping, and Integrated Electro-Mechanical Systems for Robotics
- Computer Science:
- Probabilistic Graphical Models - particularly text models and unsupervised clustering, Nonparameteric Prediction (Gaussian Processes, Support Vector Machines, Kernel Density Estimators), Manifold Learning (Isomaps, Kernel PCA, Matrix factorization, etc.), Active Learning, Consumer Preference Modeling, Information Visualization, and User Experience Research
- Design Thinking, Design Strategy, Open Innovation, and Business model generation
H2O IQ: Automated Plant Watering System
My team and I developed a wireless plant monitoring device that can automatically water plants remotely using a web-interface. One of my roles in this project was the enclosure and valve system design, along with integrating the electronics package. This involved designing for injection mold constraints, including snap fit design. We open-sourced the design, so you can download any part files here if you would like to 3D print yourself a copy!
Solar Electric Racing Boats
I led a ~20 person engineering team, called Carnegie Mellon Solar Splash, in the design of a 15ft long solar-powered racing boat. I also led the drive-train design efforts as well as most of our finite element analysis (both strength and thermal).
Fatigue and Fracture Analysis of Turbine Engines
During my time at General Electric, I worked on fatigue and fracture analysis of turbine rotor components, particularly on low pressure turbine rotors. I conducted sensitivity studies for new rotor components that we going into production, as well as analysis of broken rotors and what could be done to repair them for service.
Robotic Cooling System Design and Analysis
My team and I designed and analysed a cooling system for a robotic vechicle. We designed a radiation-shielded, heat-exchanger assembly, and created pareto plots for the performance of the device under different heat loads and environmental conditions. My role was the CAD modeling of the design along with the CFD and thermal finite element analysis.
I have done past work in topological shape optimization, which uses finite element simulations to determine the optimal amount of material necessary for a given loading condition. I have applied this knowledge to the design of an ice axe head, producing lost-weight, but high strength axe designs.
H2O IQ Printed Circuit Design
As part of the above mentioned project on a wireless plant monitoring device, one of my roles was the electronics and PCB design. This involved integrating an ATMEL controller with a solar charging circuit running off a Lithium Polymer battery, and then actuating a valve servo by reading in moisture and temperature readings.
BraveHeart: Bio-Monitoring Vests for Firefighters
I worked with a multi-disciplinary team of designers and engineers to produce a bio-monitoring vest for Mine Safety Appliances. My role was the design and prototyping of the bio-monitoring vest, which included conductive skin pads, heart rate, breathing, and temperatures sensors, and a wireless communication module capable of transmitting to a separate base unit.
Wearable Interactive Devices for Interface Control
As part of my Master's project work, I developed a 3D tracking glove input device for modeling surfaces ing 3D. This involved a lightweight micro-controller with a few pressure and button inputs that communicated via bluetooth to a larger system. You can read more about it in our paper
Crowd-generated Design Recommendation
Work I conducted while at the Palo Alto Research Center explored how to leverage crowd-given information to inform design synthesis systems. This involved coming up with new business models for crowd-based design platforms, as well as creating new algorithms for learning crowd preferences. I also implemented a functional web app for helping users customize designs.
Large-scale Clustering of Twitter Streams
My team and I implemented scalable algorithms for detecting community structure and word distributions in Twitter. My role was primarily in modeling the word distrubtions of different detected communities using topic models and empirical distributions. It turns out that Justin Beiber and other boy bands are way more popular on twitter than you could ever imagine.
Topic Models for Recipe Analysis
I have always been interested in food, and particularly as recipes as a representation of design. I decided to analyse a corpus of recipe data by using a class of unsupervised clustering models, called Topic Models, to detect clusters in recipes. In many cases the algorithm recovered groups such as "baking ingredients" or "indian ingredients". Check out some of the clusters here: LDA Clusters, HDP Clusters, hLDA clusters
3D CAD Interfaces
As part of my Master's project work, I built an Augmented Reality environment where users can put on a headset and then create objects using their hands. To accomplish this I gained familiarity with the wonderful OpenCV package, and used various vision processing algorithms to track user objects in 3D using multiple web cameras. It also involved some computational geometry work for doing surface fitting and smoothing, etc.
Collaboration Practices at Frog Design
My team and I analyzed the collaboration practices at frog Design, a global design consultancy, particularly focusing on overcoming challenges that arose during a recent corporate merger. We used on-site interviews as our primary data source and leveraged collaboration frameworks to propose recommendations for overcoming the challenges the merger imposed.