Jack Ruttan

Contact

About

I’m a computer science and math student at Western University working on AI systems, building products, and research.

Right now, I’m working at Lifemark Health Group on internal AI systems and building projects like Primate.

Hamilton, Ontario Open to summer 2027 internship opportunities Grad 2028

Experience

  1. Lifemark Health Group

    AI Systems Engineer · Technical Consultant

    Building internal AI tools to automate repetitive clinical and administrative work.

    Current

  2. Tech for Social Impact

    Project Manager

    Leading development of pro-bono software projects with the Canadian Red Cross and Brain Tumour Foundation of Canada.

    2025 — 2026

  3. Unity Health Toronto

    Applied AI Intern

    Built clinical interview agents and evaluation pipelines for healthcare research.

    Summer 2025
Move any piece

A transformer based, human like chess bot that improves by playing opponents on this site through RL. This model is currently training and it'll be playable here soon!

Follow its training on GitHub

Projects

Primate

Checks changed pages for UI bugs and posts screenshots in your pull request.

Visit Primate

Whim

Plan, shop, and play in your iMessage group chat with an AI agent.

Hack the North 2026 Best Use of Linq · Cloudflare Best Agent with a Brain

Optimate

Ranks insurance submissions by how well they fit an insurer’s risk appetite.

Hack the North 2025 Y Combinator Unicorn Prize · Federato RiskOps Gold Sponsor

Doppels

AI agents find shared interests and score compatibility through live conversations.

McHacks Best Product Design · Top 5 Finalist

Research

01
Give an LLM trusted examples 80 human-verified Comanche ↔ English translation pairs
02
Generate held-out translations The model translates 20 phrases it did not see in the examples
03
Check each generated translation Compare it with the real translation using the Levenshtein similarity algorithm
Similarity ≥ minimum threshold keep

Published in AmericasNLP 2025 @ NAACL

Advancing Uto-Aztecan Language Technologies: A Case Study on the Endangered Comanche Language

Proceedings of the Fifth Workshop on NLP for Indigenous Languages of the Americas

We compiled the first digitized Comanche dataset and a low-cost way to expand it with the goal of preserving a critically endangered language. We built a pipeline to generate translations from a small set of trusted examples, compare them with human-verified answers using normalized Levenshtein similarity, and keep only outputs that pass a quality threshold.

Read paper