Justin Bricker

Software Engineer | Data Scientist | Machine Learning

Nashville, Tennessee

View My Work

About

I'm an experienced software engineer passionate about using data, automated pipelines, and predictive analytics to solve interesting problems. Based in Nashville, Tennessee, I bring a background in computational science and mathematics to every engineering challenge.

After creating full-stack business-critical software and customer-facing applications, I believe in writing well-tested, easily extensible code and in lifelong learning through personal projects and mentoring others. My work spans data engineering, machine learning, and applied mathematics.

I'm always looking to apply my software and analytical skills to tackle new challenges and continue to grow as an engineer.

How I Work

1

Understand

Listen first. Map the problem domain, stakeholders, and constraints before writing any code.

2

Architect

Design data models, system interfaces, and pipelines with scalability and maintainability in mind.

3

Build

Iterative development with well-tested code, CI/CD, and frequent feedback loops.

4

Measure

Instrument everything. Use data to validate assumptions and quantify impact.

5

Refine

Continuous improvement through monitoring, retrospectives, and stakeholder alignment.

Experience

Senior Software Engineer, Data

AcuityMD Jan 2023 — Present

Staff Software Engineer

Eventbrite Dec 2021 — Jan 2023

Senior Software Engineer

Eventbrite Dec 2019 — Dec 2021

Machine Learning Consultant

University of Florida May 2018 — Sep 2019
  • Created multi-level machine learning models differentiating between types of Parkinsonism using diffusion MRI and clinical data gathered from an international team of neuroscientists and neurologists
  • Deployed models and output classification reports for clinicians as a tool to aid in the diagnosis and treatment of patients with Parkinsonism

Software Developer

Infinite Energy Jan 2016 — Jul 2019
  • Designed, developed, tested, and maintained business-critical internal and customer-facing applications
  • Created CI/CD pipelines using C#, React/Redux, SQL, Docker, and Kubernetes
  • Led the Agile Metrics initiative — an ETL pipeline for collecting, transforming, and visualizing data from project tracking, source control, and build tools
  • Mentored new developers through onboarding, peer programming, and code reviews

Graduate Research & Teaching Assistant

Florida State University Sep 2013 — Dec 2015
  • Research under Dr. Peter Beerli on computational genomics projects mentoring high school students
  • Taught undergraduate courses in web development and business computing

Skills

Languages

Python C# SQL JavaScript NoSQL

Data & ML

Machine Learning Data Science Data Engineering ETL Pipelines Predictive Analytics

Frameworks & Tools

React / Redux Docker Kubernetes Automated Testing CI/CD

Domains

Applied Mathematics Computational Science Full-Stack Development Medical Imaging / MRI

Projects

Selected work and contributions. More on GitHub.

AID-P: Parkinsonism Classification

Multi-level machine learning models for differentiating between types of Parkinsonism using diffusion MRI and clinical data. Deployed classification reports as a diagnostic tool for clinicians in collaboration with an international team of neuroscientists.

Python ML MRI Classification

Agile Metrics ETL Platform

End-to-end ETL pipeline for collecting, transforming, and visualizing data from project tracking, source control, and build tools. Enabled data-driven decisions to improve code quality, accuracy, and team efficiency.

C# SQL ETL Analytics

Computational Genomics Research

Research projects at Florida State University investigating downstream impacts of ambiguously condensed data formats and discovering single nucleotide polymorphisms in European water frogs using no reference genome.

Python Genomics Research Bioinformatics

Case Studies

Selected engagements with measurable outcomes.

Diagnostic ML for Parkinsonism

Challenge

Neurologists struggle to differentiate between types of Parkinsonism using clinical assessment alone, leading to misdiagnosis and delayed treatment for patients.

Solution

Built multi-level ML classification models using diffusion MRI and clinical features in collaboration with an international team of neuroscientists and neurologists.

Result

Deployed AID-P, a diagnostic tool that generates classification reports for clinicians, improving diagnostic accuracy for patients with Parkinsonism. Published findings for the research community.

Python ML Medical Imaging Research

Engineering Intelligence Platform

Challenge

Engineering leadership lacked visibility into team performance, code quality trends, and delivery efficiency across multiple systems and projects.

Solution

Designed and built an end-to-end ETL pipeline collecting data from project tracking, source control, and build tools, with automated dashboards for leadership.

Result

Enabled data-driven decisions that improved code quality, estimation accuracy, and team velocity. Adopted company-wide as the standard engineering reporting tool.

C# SQL ETL Data Visualization

Event Discovery at Scale

Challenge

A global events marketplace needed to surface relevant events to millions of users across diverse interests and locations with low latency and high accuracy.

Solution

Architected and scaled data pipelines and backend systems supporting event discovery, search, and recommendations for the platform's global user base.

Result

Systems served millions of monthly active users with sub-second response times, driving significant increases in event discovery and ticket sales.

Python Data Engineering Distributed Systems Search

Publications

AID-P: Automated Imaging Differentiation in Parkinsonism

Machine learning approach for differentiating between various types of Parkinsonism using diffusion MRI and clinical data, developed in collaboration with an international team of neuroscientists and neurologists.

Education

M.S. Computational Science

Florida State University

2013 — 2015

B.S. Mathematics

University of Florida

2009 — 2013

Things I Believe

Well-tested code is a form of documentation.

Tests express intent more reliably than comments. They prove what the code does, not just what someone hoped it would do.

Data should inform decisions, not just validate them.

Seek data before forming hypotheses. The most valuable insights come from questions you didn't think to ask.

Simple beats clever.

The best solution is the one your team can understand, maintain, and extend. Premature abstraction is a liability.

Mentoring multiplies impact.

Teaching others scales your knowledge far beyond what you can build alone. The best engineering cultures invest in growing people.

Lifelong learning is non-negotiable.

The field moves fast. Stay curious through personal projects, reading, and collaboration. Comfort is the enemy of growth.

Get In Touch

I'm always interested in hearing about new opportunities, interesting projects, or connecting with fellow engineers and scientists.