Trope: An AI platform to reliably run back-office workflows

Matthew Chow (Computer Engineering), Martin Tang (Computer Engineering), Victor Vannara (Computer Science); Hudson Hendry (Business), Jerimy Crisologo (Business) and Matthew Lim (Economics)

Degree:
- Bachelor of Applied Science
Program:
- Electrical and Computer Engineering
Campus: Vancouver

Our project

We developed Trope, an AI platform to reliably document, run, and verify workflows. Trope records a properly executed workflow across desktop and web applications and automatically converts it into documentation and live, in-app guidance. The result is faster onboarding, consistent day-to-day execution and lower support burden across teams.

Trope was developed through New Venture Design, which brings together interdisciplinary groups of students to develop and launch an entrepreneurial venture. Over eight months, we identified a high-value opportunity, built a product, secured paying customers and are in the process of seeking venture capital investment to fund the next stage of our growth.

Trope New Venture Design

Identifying the problem

We were committed from the outset to building a venture that solved a real, validated problem.

Drawing on our own work experiences and more than 30 customer discovery interviews, we identified a widespread bottleneck: organizations struggle to ensure consistent workflow adoption, change management and governance, particularly when operating on legacy software systems.

This challenge is particularly acute in the non-profit and social services sector, where staff must run high-volume government-compliant workflows on legacy case management applications. The leaders we talked to described overworked IT departments handling access requests and troubleshooting software issues, inconsistent case handling across teams, lengthy or inconsistent employee onboarding, and compliance risks.

How Trope works

Trope documents a workflow across desktop apps and web tools. It records the process end to end, including context, screenshots and user actions – capturing the true complexity of the process. Our AI agent then converts that input into structured guidance that can be accessed in real time and followed on any Mac or Windows desktop.

Today, the guidance includes step-by-step instructions, required checks and embedded context to help staff run workflows efficiently and correctly. As AI automates more digital work, Trope becomes the workflow infrastructure that helps ensure both staff and agents execute workflows reliably.

Trope workflow tool shown on a split-screen laptop interface.

How we validated our solution

There are two levels of validation to consider. The first was problem validation. The enthusiasm from the organizations we interviewed, combined with our ability to secure paid pilots, confirmed that we had identified a serious issue. The second level of validation is product validation. This will come from usage metrics, including decrease in workflow errors, reduction in onboarding dime and compliance improvements. Our initial tests suggest Trope results in two to three times faster completion and 90% fewer errors.

What we’re most proud of

Our team members have backgrounds in computer engineering, computer science, economics and business. Over the past eight months, we have learned from each other and stretched beyond our academic disciplines to gain new cross-functional skills.

In just eight months, we have launched an impactful solution that is redefining how organizations document and execute complex workflows. We’ve leveraged tools of AI coding and generation, iterated quickly, and validated immediately. Even as we were designing our user interface in Figma we joked that “if only we had Trope this would be so much easier.” Eventually, that joke became true.