Building AI-ready documentation for next-gen hardware systems - BC-986

Project type: Innovation
Desired discipline(s): Engineering - computer / electrical, Engineering, Computer science, Mathematical Sciences, Business, Social Sciences & Humanities
Company: Iris Dynamics Ltd.
Project Length: 6 months to 1 year
Preferred start date: As soon as possible.
Language requirement: English
Location(s): Victoria, BC, Canada
No. of positions: 1 - 2
Desired education level: Undergraduate/BachelorMaster'sPhD
Open to applicants registered at an institution outside of Canada: No

About the company: 

Iris Dynamics Ltd. develops and manufactures advanced smart linear motors that integrate drivers, force and position sensors, and controllers into rugged, high-performance units. Our motors are uniquely suited for applications requiring precise force control, smooth motion, compliance and high durability - such as industrial automation, robotics, and harsh-environment systems. By combining fully integrated electronics, Iris Dynamics’ technology offers superior efficiency, simplified installation and reduced system complexity compared to traditional actuator solutions. Iris Dynamics is positioned to support major scale-up of production and commercialization across multiple sectors, aligning with Canadian excellence in advanced manufacturing, and offering strong potential for collaboration, research partnerships, and value-added innovation.

Describe the project.: 

Iris Dynamics Ltd. manufactures fully integrated linear actuators with onboard control, sensing and logic. We are seeing a growing number of customers using large-language models (LLMs) to discover and integrate our devices into their applications. Our documentation and data infrastructure, however, were not originally built with LLM ingestion or AI-augmented workflows in mind. This project will develop a student-driven team (funded via Mitacs) to validate, restructure and augment our public documentation and developer workflow so it becomes “AI-ready” and decreases friction for customers.

Key activities include: testing how major LLMs (e.g., ChatGPT, Claude) interpret our devices, run customer-use scenarios (e.g., force/displacement monitoring, mechanical backlash detection, software/firmware setup) from LLM-only instructions or public docs, identifying failure modes or misleading guidance, flagging documentation gaps, and then delivering a structured export (Markdown/JSON) of “AI-user” instruction sets. A secondary stream will build the internal foundation for AI-enabled operations: modular APIs, secure data indexing of CRM/support logs, and a documented framework for future agents and automation. Ownership remains with Iris; students work under supervision and outputs become part of our long-term knowledge base.

Expected outcomes: faster onboarding, fewer support tickets, higher visibility in AI-augmented development environments, and a documented process for ongoing LLM-testing and documentation refresh. Phase 1 (public docs & LLM validation) is prioritized, with internal AI tooling as a parallel or follow-on phase.

Required expertise/skills: 

Students should have strong practical and analytical skills in applied AI, software engineering, or mechatronics. Core technical skills include:
• Programming: Python (preferred), C/C++, and basic web technologies (HTML/Markdown/JSON).
• AI/ML tools: Familiarity with large-language-model APIs (OpenAI, Anthropic, Gemini, etc.) and retrieval or vector-database frameworks (LangChain, LlamaIndex, FAISS, Pinecone).
• Data management: Experience structuring and tagging technical documentation; understanding of metadata, schemas, and information retrieval.
• Embedded or controls background: Ability to test and operate electromechanical systems, interface sensors/actuators, and log performance data.
• Version control and documentation: Git/GitHub proficiency, technical writing, and clear reporting.

Assets (optional)
• Understanding of product management or customer-development processes.
• Experience with CRM or ERP systems (HubSpot, Salesforce) and how they link to sales/support workflows.
• Familiarity with prompt engineering, data governance, or AI ethics.
• Awareness of marketing and documentation practices that improve discoverability (SEO, “LLM SEO”).

Ideal candidates bridge technical and commercial thinking — able to test, document, and interpret how AI-enabled workflows improve customer experience and business efficiency.