• Led adoption of Git for version control, including configuration of a private server for internal repos and staff training.• Modified ANTs code to work with SLURM-based HPC clusters (DEAC & DEMON) and contributed code for inclusion in ANTs open source product, which is now utilized by SLURM users worldwide.Contributed to various software solutions that included:• Software for automated, large-scale assessment of muscle attenuation and cross-sectional area using computed tomography (CT) imagery.• Software that mimics DEAC job arrays in DEMON to allow researchers to migrate jobs from the DEAC HPC cluster to the DEMON HPC cluster, significantly reducing turnaround time and improving research innovation.• A tool to analyze unstructured web server log files and recover research data for a study that had lost its primary research records. Used Bash, MySQL, Python, SQLAlchemy, Aqua Data Studio, and DataGrip.• A publication visualization system for enhanced dissemination of results in the LIFE project, built in Python, JavaScript, NPM, D3, HTML/CSS, NGINX, Flask, et al.• A Zabbix-based monitoring system to proactively identify service degradation on dozens of production servers.• Software that performs automated feature extraction on fundus imagery to predict future health problems, built using Java, Smile, ImageJ, Apache Commons (Math, Collections, CSV), Guava, TestNG, and Mockito.• An ANTs-based image processing pipeline to analyze thousands of brain tissue MRIs in support of novel research into the effects of air pollution on cognitive decline.

Contributed to various software solutions that included:• An advanced automatic crash notification algorithm to improve triage decision-making for 150K+ injured in car accidents annually.• Novel web-based assessment instruments to collect perspectives on injury severity using Java, Liquibase, Spring, GWT, AWS, Guava, and Velocity. The data collected allows us to relate ICD-9 codes to their new ICD-10 counterparts.• An extensible Flex-based framework for clinical evaluations of the cognitive functioning of older adults.• Multiple iOS and Android apps for clinical research and partners.• The Cluster Jobset System, a software package designed to facilitate exploitation of underused high-performance computing cluster resources, thus permitting faster research iteration.• Tools for automated manipulation of LS-Dyna input and output data, which allows researchers to focus on expert analysis instead of error-prone manual operations.

• Introduced cloud computing and Amazon Web Services for flexible, high-performance, resource-intensive computation.• Mentored an intern team building an algorithm to construct a 3D surface mesh from MRI-produced point cloud data.• Supervised and led development activity for a distributed team of four.Contributed to software solutions that included:• Database/data management software to support statistical analysis of heterogeneous hospital discharge databases containing 80M+ records.• An interactive, video-based, IRT-backed assessment tool used in clinical contexts, built with Java, JasperReports, Spring, Cortado, and other libraries.• Multi-platform cluster computing solutions to facilitate and accelerate research performed by non-programmers.• A platform built in R and C# to automate the generation of comprehensive data sets for simulations, dramatically increasing cohesion of data and documentation and decreasing turnaround time to generate simulation data.

Instructed one section each of the course titled Computer Science II (CSC230) and the course titled Algorithm Analysis (CSC331). CSC230 uses Java to explore object-oriented design patterns, inheritance, the Java Collections Framework, generics, and software design best practices and tools such as JUnit and log4j. CSC331 is a language-agnostic analytical survey of common algorithms and algorithmic techniques, emphasizing efficiency and correctness.

Instructed three lecture and laboratory sections of the course titled Introduction to Computer Science (CSC111) and four laboratory sections of the course titled Overview of Computer Science (CSC101). CSC111 introduces students with no programming experience to Java and emphasizes algorithmic problem solving, data types, branching and looping constructs, arrays, collections, and basic searching and sorting algorithms. CSC101 develops an understanding of fundamental computing concepts among students with minimal computer experience via hands-on experience with HTML, CSS, multimedia software, and database tools.

Conducted research for compact multi-lens diversified imaging system under a grant from IARPA. Independently investigated, implemented, and evaluated both simulated and real-world solutions to project obstacles, provided assistance with performance evaluation tasks, and conducted time-saving literature reviews. Designed and implemented an extensible, camera-independent software interface for demonstrating and interacting with optical and computational tools.Explored algorithmic solutions to the problem of reconstructing information "hidden" in the sub-pixel displacement of images produced by multiple lenses in an array. Developed software to simulate the effects of noise, blurring, and inaccurate reconstruction parameters. Simulated results were comparable to empirical data, indicating that, based on several metrics, images can be reconstructed with reasonable fidelity even under "real world" conditions. Funded by grant from the Disruptive Technologies Office. Presented research at 2006 SPIE Optics & Photonics Conference.