Core contributor to Shopify’s AI agent platform that powers Sidekick (Shopify’s AI assistant for merchants) and other agentic experiences at Shopify. I’m focused on improving the performance, quality, and capabilities of the platform, working at every layer from the language runtime itself to the web server, load balancer, and fine-tuned models, as well as collaborating directly with major LLM vendors.

Taught a section of CS 350 - Operating Systems, a required third-year Computer Science course, for the Fall 2024 term. Devised lecture materials, assignments, and examinations in collaboration with the instructional team.

Developed tools and infrastructure for our lab’s research projects, and helped integrate our research into products. Built the initial prototype of SmartFind, which uses deep learning to improve the recall and question answering capabilities of the Microsoft Word search box. Contributed to the Hybrid Reward Architecture paper—the first AI technique to beat the Atari 2600 game Ms. Pac-Man, and TextWorld—an RL environment for text-based games.

Led the development of our virtual personal assistant framework, using state-of-the-art natural language processing techniques. This product was deployed to millions of customers through partnerships with leading smartphone and smart TV makers. In 2017, Maluuba was acquired by Microsoft, and became a new Microsoft Research lab specializing in deep learning, language understanding, dialogue systems, and reinforcement learning.

Worked on the Drupal front-end of UNITIS, an information system that manages people, groups, and assets for departments and faculties at the university. UNITIS powered the campus-wide staff directory, as well as many department webpages including the Department of Mathematics and Statistics.

The Aurora Single Level Store Operating System
Emil Tsalapatis, Ryan Hancock, Tavian Barnes, Ali José Mashtizadeh
2021. In Proceedings of the ACM SIGOPS 28th Symposium on Operating Systems Principles. Association for Computing Machinery, New York, NY, USA, 788–803.

TextWorld: A Learning Environment for Text-based Games
Marc-Alexandre Côté, Ákos Kádár, Xingdi Yuan, Ben Kybartas, Tavian Barnes, Emery Fine, Jasmine Moore, Ruo Yu Tao, Matthew Hausknecht, Layla El Asri, Mahmoud Adada, Wendy Tay, Adam Trischler
2018. In Computer Games, 7th Workshop, CGW 2018, Held in Conjunction with the 27th International Conference on Artificial Intelligence, IJCAI 2018, Stockholm, Sweden, July 13, 2018, Revised Selected Papers. Springer Nature Switzerland, Cham, Switzerland, 41-75.

Hybrid Reward Architecture for Reinforcement Learning
Harm van Seijen, Mehdi Fatemi, Joshua Romoff, Romain Laroche, Tavian Barnes, Jeffrey Tsang
2017. In Proceedings of the 31st International Conference on Neural Information Processing Systems. Curran Associates Inc., Red Hook, NY, USA, 5398–5408.

Method and system for classifying queries (US10467259B2)
Kaheer Suleman, Jing He, Tavian Barnes
2015