Software/Toolkits
Textalytic
Browser based text analysis that handles pre-processing, analyzing, and
visualization in an easy to use web interface. (Written in vanilla JavaScript, charts/graphs use
D3.js & Chart.js, table uses Handsontable). Application is broken
out into (5) node.js microservices.
Stylometry Toolkit
A browser based toolkit to identify stylistic signatures characteristic of
Latin prose and verse using a combination of quantitative stylometry and supervised machine
learning.
Anagram Toolkit
A browser based sequence alignment toolkit for the detection of anagrams in
Latin literature.
Filum Toolkit
A tool for identifying verbal resemblances in literature written in JavaScript
that uses FuzzySearch & Levenshtein Distance.
Work Experience
RapidVerify - Founder
- Engineered a user-friendly interface enabling businesses to deploy robust verification processes without requiring extensive technical expertise.
- Employed advanced algorithms and data analytics to ensure accurate, real-time verification, reducing the risk of fraudulent activities.
- Integrated the platform with leading CRM, property management systems, and HR platforms, including Greenhouse, Lever, QuickBooks, DoorLoop, Pipedrive, and HubSpot, facilitating a broader adoption and enhancing user experience.
- Implemented end-to-end data encryption and comprehensive audit trails, successfully meeting compliance standards and ensuring the highest level of data security and privacy for sensitive user information.
- Ensured high availability and platform resilience, achieving 99.99% uptime by architecting and managing the infrastructure on Kubernetes, coupled with robust DevOps automation using Flux for GitOps-driven continuous deployment and automated recovery processes.
LitX - Sr. Full-Stack Engineer
- Led the development of a transformative MVP for legal sector information exchange.
- Spearheaded the architectural design of a robust database schema
- Pioneered the implementation of resumable downloads, enhancing user experience significantly.
- Expertly developed and integrated REST endpoints, coupled with constructing a dynamic front-end using React.
- Instrumented Kubernetes for improved deployment efficiency.
- Documented the entire process meticulously in Confluence, ensuring knowledge transfer and sustainability.
CapX - Sr. Software Engineer
- Build out the internal tool used by employees in React and integrate with the backend.
- Migrate our CI/CD workflow from Jenkins to Github Actions
- Redefined database schema to return smaller responses to the frontend
- Prototype Excel extraction from Financial Statements
- Successfully reduced MongoDB query time by 60% through the strategic use of aggregation pipelines.
- Developed an automated scraper using Puppeteer to source email leads.
Under Armour (projekt202) - Sr. Developer
- Convert Monolithic Python Service to Modern Spring Boot Microservices
architecture.
- Implement CI/CD with Jenkins, Docker, & Kubernetes.
- Support React.js UI Framework
for Brand Challenges.
- Utilize tools such as Celery, Rabbitmq, Python, MongoDB, Prometheus, Kubernetes,
Confluence, jira, and bitbucket.
- Meet with product & design teams for Sprint planning & retros.
- Meet with technical leadership for high-level architecture choices.
- Lead team of frontend, backend developers, Android developers, & iOS developers.
- Implement external APIS such as Segment, Snowflake, Amplitude, & Sales Force Marketing
Cloud
UT Austin - QCL Lab - UI/UX & JavaScript Developer
- Research literature using machine learning, natural language processing,
bioinformatics, and systems biology.
- Implement automated CI/CD with Gitlab & Docker
-
Re-write filum tool which uses a technique derived from computational
biology known as sequence alignment, which considers the character-by-character similarity of
phrases in JavaScript.
- Collaborate with the web team on UI/UX.
- Automate large literature text file
analysis.
Austin Software Works - Full Stack Developer & ML
Engineer
- Proficient in PHP/MySQL | JavaScript/MongoDB | Meteor/Nodejs
- Develop/Debug/Test custom
applications for business clients
- Use R for data analysis and hypothesis testing
-
Deploy application code to AWS EC2/RDS
Instances
- Commit code changes to Github via Git
- Create visual charts using D3 & tableau
-
Create Tensorflow Models in Python
- Optimize NVIDIA CUDA Code & OpenMP for GPU computing.
UT Austin - Project Manager
- Oversee technicians for VoIP Conversion of 20k phones
- Modify/Configure Cisco Switches to
accommodate new VoIP Phones
- Converted spreadsheet to efficient PHP/MySQL Data Entry
Workflow
- Reduced survey time by 50%
& deployments by 30%
- Assist in planning of best practices for site surveys
- Ensure
surveys and deployments ran on schedule
- Migrate site surveys from Excel to a web based
application
The I.T. Department - Head Geek
- Install & Configure CentOS/Ubuntu AWS Instances
- Full-Service I.T. Provider to local
Austin small businesses
- Move physical servers to hosted solutions in AWS
- Address trouble tickets opened by
customers via email
- Utilized tools such as VMWare, Git, HeidiSQL, Putty
- Install &
Configure Cisco Routers, Switches & Access
Points
- Implemented VOIP network using Cisco Appliances
- Configure and Manage Load Balancers
and SSL Encryption Devices
- Facilitate communication and develop relationships for many cross functional teams
SolarWinds - Sales Engineer
- Determine the strategic needs of IT management for enterprises.
- Assist customers in
designing network/storage system architecture.
- Demonstrate expert knowledge of vendor
specific technologies including VMWare, Citrix,
and Cisco.
G.S.A. - Computer Technician
- Deploy new Dell laptops to all of members of the GSA
- Backup all user data to fiber
networked NAS drives
- Ensure encryption was set on all user machines
- Ensure computers could log onto
encrypted network
- Ensure all user data was restored
Research
Estimation of potential United States influenza mortality. An agent based model
simulation
Planning of the next influenza pandemic is of vital concern to many public health officials. The
aim is to use an agent based model of influenza in the United States to estimate mortality
should a pandemic occur present day. Models of infectious disease
epidemics have been proven useful in understanding the dynamics of disease transmission,
vaccination strategies, and are constantly used as an apparatus to support decisions made by
public health officials. Using an agent-based
model of Influenza A virus infection, a simulation is created to simulate the effects of two
strains of influenza A causing a pandemic in the United States.
Filum: Large-scale identification of literary intertextuality using sequence
alignment
Profiling of intertexutal relationships is foundational for literary study. We demonstrate
Filum, a user-friendly tool that employs character-level sequence alignment to detect verbal
parallels with or without lexical overlap. When applied to a database
of more than 1,000 intertextual parallels of known significance from the Latin poet Valerius
Flaccus, Filum is able to recover more than 80% with reasonable specificity and to identify more
than 250 new intertexts previously unrecorded
in the scholarship.
Computational Differentiation of Latin Literary Genre
This article describes a new quantitative approach to the study of Latin literary genre. Using
computational text analysis and supervised machine learning we construct a detailed stylistic
profile of the vast majority of extant classical Latin literature
and classify works by traditional genre with high accuracy. By examining the statistical basis
for these automated classification decisions, we identify salient stylistic characteristics of
each genre at the level of syntax and
non-content vocabulary. Through a series of case studies, we illustrate how this approach
enables both confirmation at scale of long-appreciated stylistic tendencies and identification
of unrecognized generic signatures.
A Stylometry Toolkit for Latin Literature
Computational stylometry has become an increasingly important aspect of literary criticism, but
many humanists lack the technical expertise or language-specific NLP resources required to
exploit computational methods. We demonstrate a stylometry toolkit
for analysis of Latin literary texts, which is freely available at www.qcrit.org/stylometry. Our
toolkit generates data for a diverse range of literary features and has an intuitive
point-and-click interface. The features included
have proven effective for multiple literary studies and are calculated using custom heuristics
without the need for syntactic parsing. As such, the toolkit models one approach to the
user-friendly generation of stylometric data,
which could be extended to other premodern and non-English languages underserved by standard NLP
resources.
Presentations
How to Do Philology with Computers - Boston, MA
Computational stylometry has aided the work of philologists for over 50 years. From simple word
counts to the latest use of machine learning for authorship attribution, computation offers the
literary critic a wide array of techniques to better understand
individual texts and large corpora. To date, these methods have largely been accessible to
specialists possessing a background in programming and statistics. The Quantitative Criticism
Lab has now designed a user-friendly toolkit
that will allow humanists with no prior training in the digital humanities to obtain a wide
range of philological data about most classical texts and to perform sophisticated quantitative
analyses—all using a simple point-and-click
interface. This presentation will demonstrate some of the experiments and literary critical
insights enabled by the toolkit, and discuss relevant issues of interpretation and statistical
analysis.
