I am a Principal Member of Technical Staff at Sandia National Laboratories in Albuquerque, New Mexico. My work focusses on understanding and evaluating the credibility of computational modeling and simulation (ModSim) through verification, validation, and uncertainty quantification (V&V/UQ).
I did my undergrad at Carnegie Mellon in the physics department and then started my graduate work at Johns Hopkins in mechanical engineering. I completed a masters there and then moved (with my advisor) to Duke where I completed my PhD in 2015. My dissertation can be found in the Duke library.
My professional interests are in the field of modeling and simulation with a focus on uncertainty quantification (UQ), verification & validation (V&V), and general credibility assessment.
My main research and practical interests are in data-driven methods to perform sensitivity analysis and uncertainty quantification. In particular, I am interested in pragmatic approaches that balance sophistication with need and cost.
My graduate work focussed on Polynomial Chaos Expansions but I have since become an advocate of using Kriging methods. I am also a big fan of empirical analysis to avoid the illusion of precision that surrogate models tend to provide.
I also do a good deal of Python development for tools in this area and hope to eventually open source my ever-growing toolbox.
I do some research but most of my focus is on applications. Any research directly comes from trying to solve a direct problem.
While a physics major at Carnegie Mellon, it was drilled into us that you must document your process in order to improve.
As such, I began to journal everything I cook (and later, everything my wife cooks too). It began as a WordPress blog but I eventually moved to my own bespoke CMS (more on that later). When possible, I try to include any notes and changes as well as a photo and the original recipe. Sometimes, it is just a picture and a link.
I also do a fair bit of Python development both professionally and personally. Some, though not all, is available in my Github Profile
Some notable projects include:
I've written two major file synchronization tools (and some deprecated ones). PyFiSync and syncrclone.
PyFiSync was originally designed to synchronize over rsync and made a valiant effort to track files even when modified. It is fairly performant (thanks in part to my other project DictTable) and offers many features. I later added rclone remote support
syncrclone was born out of the desire to be able to sync two rclone remotes and to simplify the synchronization algorithm (as rclone always fully re-copies the files). It too uses DictTable internally and can handle very large repositories
I am a complete novice with front-end development but I have designed my own CMS to drive KitchenKatalog called NBweb. It is rough but has enabled me to be fully in control! This is actually the 3rd iteration on the software for KitchenKatalog but the first to be dynamically built. It works well for my uses but the code desperately needs to be refactored.
Every page is simply Markdown so that they are both future-proof and easy to manage. I also designed my own search algorithm tailored to the type of queries I perform.
I also built PhotoStory out the desire to extend the photo capabilities in NBweb. It follows a similar philosophy but is much cleaner code.
I am also beginning to dabble in web-facing system administration. I haven't decided if I enjoy it or if it is more of a chore. But if you're able to read this, then it is currently working!
Other personal projects include notefile, a tool to create and manage sidecar datafiles.
I have also developed a fast in-memory data-object called DictTable. It began as a "Yak-Shave" for file synchronization but I have come to develop it separately.
Last Updated: 2022-12-20T135859-0700