Case Study: Creating an AI Solution to Tailor Learning Paths

This case study outlines how I led the learning design efforts to create an AI-powered feature that automatically generates personalized learning paths.

generate an image about a team working c

Introduction

As a Senior Learning Designer, I was responsible for developing the Learning Requirements Document (LRD), creating research whiteboards, and collaborating with cross-functional teams, including product, design, engineering, and data science. My role involved providing design feedback, guiding the learning design process, and ensuring that the AI-powered learning paths were aligned with learning science research and learning design principles.

The Challenge

The challenge was to create an AI-powered solution that could generate coherent and effective learning paths tailored to the specific skill needs of individual learners. The AI system needed to create personalized skill trees and then use them to generate aligned learning paths that were personalized, relevant, and adaptive, ensuring that learners progressed efficiently through the material.

Key issues included:

Scalability: Developing an AI solution that could scale to accommodate a wide range of learners and learning scenarios.
Relevance: Ensuring that the AI-generated learning paths were aligned with individual learner needs, helping them fill knowledge gaps and build new skills effectively.

a puzzle that is simple with 4 pieces.jpg

The Process

I took the following steps to complete this project:

Research & Discovery: I began by conducting in-depth research on skill trees and learning paths, creating a research whiteboard to visualize key concepts, dependencies, and best practices. This research helped identify how AI could be used to map learners’ existing skills to the necessary steps for their development.

Learning Requirements Documentation (LRD): I wrote two LRDs to define the project’s learning design aspects:
- Skill Trees: Mapping the skills learners need and understanding their relationships and dependencies.
- Learning Path Generation: Designing how the AI could leverage skill tree data to generate personalized learning paths based on the learner’s current skill level.
Collaboration & Iteration: Throughout the development process, I:
- Held weekly sessions with the Product Manager (PM) and regular check-ins with the PM and design teams to discuss progress and clarify requirements.
- Worked closely with the data science team to ensure that the AI algorithms were grounded in learning science, and aligned with the objectives of skill development and knowledge acquisition.
- Participated in consultations, providing learning design insights and feedback on AI outputs, helping shape the AI’s decision-making process.
Content Support & Testing: I contributed to the creation of user support materials, writing three articles that helped instructors understand how to use the new AI feature effectively. I also collected feedback from the marketing team to refine these articles further.

I tested the feature in its alpha and beta stages, evaluating the AI-generated learning paths and providing valuable feedback to the team. My feedback helped refine the AI’s logic and ensure that the final product aligned with user needs.

The Solution

The result was an AI-powered feature that could automatically generate personalized learning paths for learners. This system used skill trees to assess learners’ current knowledge and tailor the learning journey to fill gaps and ensure relevant progression. The feature was intuitive for administrators to use, making it easy for them to create and manage personalized learning paths for learners at scale.

For additional information, see this press release and support article.

create an image to show a learner who is

Results, Impact & Reflection

The introduction of the AI-powered learning path feature led to a 30% Increase in Admins creating learning paths. Through feedback, admin reported that this feature made it significantly easier for them to create customized skill trees and learning paths, increasing engagement and usage.

Reflection:

This was my first experience working on an AI project, and it was a valuable learning opportunity. I learned how to collaborate with the data science team to ensure that AI-driven solutions are grounded in sound learning science principles. The project also taught me the importance of continuous iteration and testing, as well as how to balance the technical capabilities of AI with the needs of learners.