Skip to content
Jamiu AbdulfataiProduct Designer
← All projectsView visual case study

PRODUCT DESIGN CASE STUDY

Chalant AI

Chalant AI project overview

MY ROLE · Learner and administrator product design

Chalant AI

I designed Chalant AI to make learning about AI feel like working with it. The AI Manager Path gives learners a team of agents, tasks to assign and outputs they need to judge.

01What I wanted to solve

I worked across the learner workspace, review queue, SOP library and export tools, alongside the administrator’s course and analytics views. The challenge was keeping human judgement part of the workflow.

02The decisions that mattered

Make the lesson a workspace.

I used agent roles, status and tasks to turn learning into decisions made in context.

Put review into the task flow.

I gave Needs Review its own stage and brought confidence and SOP-compliance signals into the review queue.

Give quality a reference point.

I connected work to versioned procedures, so the learner has something more concrete than personal preference to review against.

03See the work

The screens below show how those decisions came together.

The AI Agent Workspace: five agents with role, live status, current task and progress, alongside accuracy and volume figures, the learner reads their team the way a manager reads a standup
The AI Agent Workspace: five agents with role, live status, current task and progress, alongside accuracy and volume figures, the learner reads their team the way a manager reads a standup
More screens and supporting work
The Review Queue states each output's confidence score, an estimated review time and whether it passed SOP compliance, Quick Approve sits beside a full Review rather than replacing it
The Review Queue states each output's confidence score, an estimated review time and whether it passed SOP compliance, Quick Approve sits beside a full Review rather than replacing it
The SOP Library: versioned standard operating procedures with compliance rings, a critical alert surfaced at the top, and each SOP linkable directly to a task
The SOP Library: versioned standard operating procedures with compliance rings, a critical alert surfaced at the top, and each SOP linkable directly to a task
The Export Center turns practice into artefacts, agent reports, analytics, decks and task data leaving in PDF, Excel, PowerPoint and JSON
The Export Center turns practice into artefacts, agent reports, analytics, decks and task data leaving in PDF, Excel, PowerPoint and JSON
Admin analytics where each AI insight names the course and module, quantifies the drop-off, recommends an action and states its own confidence
Admin analytics where each AI insight names the course and module, quantifies the drop-off, recommends an action and states its own confidence

04What I can show, and what I would test

The design artifacts show the review structure and shared learner/admin system. I would test whether learners become better at spotting unsuitable output across sessions; I have not measured that learning outcome.

View chalantwork.com All projects