Introduction
An EdTech company building a corporate learning platform wanted to move frontline sales and service training beyond static, click-through slides. RethinkingWeb built an AI roleplay module that lets learners practise real conversations with an AI character, scores each attempt against a trainer-set rubric, and writes the results straight back into the LMS as course progress.
The result is a training experience learners actually engage with, and a dashboard trainers can use to see exactly who needs coaching next.
Problem Statement
Slide-based e-learning wasn't building the conversational skills frontline teams needed on the job — reps could click through a module without ever practising how to actually handle a real customer conversation.
Trainers also had no scalable way to run consistent roleplay practice across large, distributed teams, and no easy view into which reps were falling behind or had gone quiet, until it was too late to act.
Solutions
RethinkingWeb built an AI-powered roleplay engine embedded directly inside the LMS. Key elements included:
AI Character Roleplay Engine
The AI character stays in role for the whole scenario, adapting to whatever the learner says.
Trainer-Set Rubric Scoring
Every attempt is scored automatically against a rubric the trainer defines, not a generic checklist.
LMS Progress Sync
Roleplay results are written straight back into the LMS as course progress, with no manual entry.
Team Performance Dashboard
Trainers see average scores, sub-skills on track, and reps at risk across the whole team at a glance.
Needs-Attention Flagging
Reps with no recent activity or open coaching flags are automatically surfaced for follow-up.
AI Team Assistant
An in-dashboard AI assistant answers trainer questions about any rep's performance in plain language.
Implementation
The engagement covered the AI roleplay conversation engine, rubric-based scoring logic, LMS integration to sync results as course progress, and the team performance dashboard — spanning overview, product knowledge, skill heatmap, roster and engagement views — plus the in-app AI assistant trainers use to query rep performance.
Results & Achievements
The platform came away with a training experience that actually builds conversational skill, and full visibility into how teams are progressing:
Reps Practise Real Conversations, Not Slides
Learners talk through scenarios with an AI character instead of clicking through content.
Consistent Scoring Against a Shared Rubric
Every attempt is judged against the same trainer-defined standard, across every rep and team.
At-Risk Reps Surfaced Automatically
Inactive or struggling reps are flagged for trainers before they fall further behind.
Trainers Get Instant Answers
The AI assistant answers performance questions in the dashboard, with no manual digging.
Conclusion
By turning static slides into AI-driven roleplay practice, RethinkingWeb gave this learning platform a training experience reps engage with and a dashboard trainers can act on — with every result flowing straight back into the LMS.
