Justin Harris is an advisor and senior engineering leader with 13+ years building production AI systems and platforms, currently shaping Sidekick at Shopify and advising Tahoma AI while running Pill0’s technology as CTO. He combines deep ML/NLP experience from Maluuba and Microsoft (Copilot, Bing Chat, Copilot Studio) with hands-on full‑stack and infrastructure work—authoring Rust services with Python/TypeScript/Ruby bindings and shipping observability and plugin frameworks used at scale. Justin is a pragmatic builder who moves research into production, having optimized memory- and CI-driven workflows in notable open-source projects like Maluuba’s nlg-eval and contributed to graph and blockchain demos. He favors clarity, testability, and reuse, and brings a rare mix of systems engineering, research fluency, and product intuition to healthcare and commerce-focused AI products. Based in Toronto, he holds a double honours BMath from University of Waterloo and often surfaces non-obvious operational improvements that make large ML systems reliably experimentable.
12 years of coding experience
13 years of employment as a software developer
Diploma of College Studies Honours Science, Diploma of College Studies Honours Science at CEGEP - John Abbott College
BMath double major Honours Computer Science and Honours Combinatorics & Optimization, BMath double major Honours Computer Science and Honours Combinatorics & Optimization at University of Waterloo
Contributions:3 releases, 65 reviews, 117 commits in 3 years 6 months
Contributions summary:Justin added a demo and simulation code for the "Sharing Updatable Models (SUM) on Blockchain" project. The changes included the implementation of a model component, likely for the front-end, incorporating elements like a UI for data input and display. The code also interacts with backend contracts using web3, which indicates back-end integration. These changes indicate a focus on implementing a demo of the project with the addition of front and back end functionality.
Evaluation code for various unsupervised automated metrics for Natural Language Generation.
Role in this project:
MLOps Engineer
Contributions:2 reviews, 24 commits, 38 PRs in 2 years 9 months
Contributions summary:Justin focused on setting up and configuring the build and test environment for the nlg-eval project, integrating Travis CI for automated builds and tests. They modified the `setup.sh` script to allow re-running and prevent re-downloading of files, improving the setup process. They addressed Python 2 compatibility issues and improved the memory usage of the project by making some changes to the tests and also by improving the code to remove old metrics. The user also introduced changes to the meteor file to add a memory check, and made several optimizations to code to reduce the memory.
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.