Engagement bots | Byju's

Problem Statement

Byju's sought to leverage AI and machine learning to improve student engagement, personalization, and overall learning experience within their virtual classrooms.

Solution

• Developed a framework to identify key AI bot roles for increased student engagement.
• Prioritized initial use-cases and designed interaction flows and UI for a mascot-based bot.

My Role

• Led the UX design for the project.
• Conducted research and facilitated brainstorm sessions.
• Created the AI bot framework.
• Collaborated with product managers to prioritize features.
• Led the art direction for "Sparky", the mascot-based bot.

Design Process

Research & Ideation:
• Brainstorm sessions with the team to explore AI's potential within Byju's class sessions.
• Analyzed ideas and identified three key bot roles.

Framework Development:
• Created a framework for AI bot intervention, providing clarity for future use-case identification.

Use-Case Prioritization:
• Used effort vs. impact matrix with product managers to select initial use-cases.

Bot Design & UI:
• Art directed "Sparky" – a mascot bot with animated loops to enhance engagement.
• Designed interaction flows for FTUE and doubt-resolution use-cases.

Initial Use-Cases

FTUE (First-Time User Experience): Sparky welcomes and guides students through onboarding.

Doubt Resolution: Sparky encourages students to ask questions and directs them to the class forum.

Polls: Sparky prompts poll participation for increased interactivity.

Outcomes

Strategic AI Integration: The framework provides a clear blueprint for future AI enhancements.

Increased Engagement: Sparky, with its friendly animations, aims to boost student interaction and enthusiasm.

Data-Driven Refinement: Staging trials will gather insights to fine-tune Sparky's interactions for optimal impact.

Brainstorming and ideation based on the phases of the class session and the core design outcomes that we have envisioned for all design projects.

An exhaustive set of ideas were noted down with the help of the team.

After analysing the ideas, it became evident that each fall into one of the mainly three roles that a bot could play in the class and interventions could be looked through that lens and implemented.

Prioritisation of the use-cases was done using effort vs impact matrix along with the product managers

Planned and art directed the character design of a mascot based bot with the help of a character designer.

We designed a number of animated loops of this mascot bot character called 'Sparky'.

Initially the bot is being experimented in the FTUE use-case.

Another use-case that we are doing is to aid in doubt resolution.

Sparky points to the class forum, encouraging the student to ask any doubt he has

Sparky comes in polls as a prompt to participate

The set of initial use-cases are developed and currently on staging using the above shown animated loops. After the trial run is done with free users, Sparky experience will be fine tuned and integrated.

You may also like