Case Study: iQMA — AI-Powered Smart Training Assistant

The Problem

The training industry has a dirty secret: the ROI of most professional development programmes evaporates within days of the final session. Participants leave with energy and good intentions — and return to exactly the same habits the following Monday.

🧠

The forgetting curve

Without reinforcement, learners forget up to 70% of new information within 24 hours and up to 90% within a week. A single workshop cannot overcome this biology.

🔁

No practice environment

After the session ends, there is nowhere for learners to safely practise new skills — especially AI skills that need contextual, immediate feedback to embed properly.

📊

No visibility for managers

Training sponsors have no way to track whether skills are being applied — leaving L&D investment as an article of faith rather than a measurable business outcome.

🎯

Generic reinforcement doesn't work

Off-the-shelf LMS modules don't reference the actual documents, tools, or workflows from the training session — making reinforcement feel disconnected from real work.

A workshop that isn't reinforced is a cost, not an investment. iQMA turns the cost into a compounding capability.

The Solution — 4 Features

iQMA is a document-grounded AI coaching platform. Trainers upload the materials from a session — course notes, SOPs, case studies, reference guides — and iQMA builds a contextual AI coach that speaks the learner's actual language, references their actual content, and practises their actual workflows.

01

Knowledge Base — Upload Your World

Trainers upload their training materials — PDFs, DOCX, PPTX, or TXT — and iQMA ingests them as the grounding layer for everything that follows. The AI coach doesn't answer from generic internet knowledge; it answers from your specific documents. Every response is anchored to what was actually taught.

Knowledge Base — Document Upload Live UI

The iQMA Knowledge Base interface — trainers upload course materials (PDF, DOCX, PPTX) which become the grounding layer for the AI coach and all practice exercises.

The iQMA Knowledge Base interface — trainers upload course materials which become the grounding layer for the AI coach and all practice exercises.

02

Practice Exercises — AI Coach in Context

Trainers create guided practice exercises mapped to specific skills or scenarios from the session. Each exercise gives the AI coach a scenario, a set of learning objectives, and a structured approach — so the learner gets a conversation that mirrors a skilled human coach asking the right questions at the right moment, not a static quiz or a generic chatbot.

Create Practice Exercise — Scenario Builder Live UI

The Practice Exercise creation interface — trainers define what the team practises, why it matters, and the scenario the AI coach uses to guide the learning conversation.

03

Training Assistant — Live AI Coaching

Core Feature

The learner-facing Training Assistant is where capability is actually built. The AI coach guides each learner through the exercise step by step — asking questions, diagnosing gaps, providing feedback, and tracking progress in real time. The coach knows the documents. It knows the scenario. It responds like a subject matter expert who has read everything the learner was taught.

Training Assistant — AI Coaching Conversation

The learner-facing Training Assistant — progress tracking, AI coaching conversation, and practice exercise navigation in a single interface.

04

Knowledge Quiz — Competency Verification

Beyond practice, iQMA generates knowledge quizzes auto-derived from the uploaded documents — not hand-crafted by the trainer. Questions test genuine understanding of the material taught, giving both learners and managers a verifiable competency signal rather than a completion certificate.

Knowledge Quiz — Auto-Generated from Documents

The Knowledge Quiz interface — questions generated directly from uploaded training materials. Learners see the source document reference alongside each question, grounding assessment in what was actually taught.

How iQMA Fits Into a Programme

iQMA is not a standalone platform — it is the reinforcement layer that makes any training investment compound over time. It is offered as a follow-on to EdTech Revolution workshops and engagements, or as a standalone add-on for organisations with existing training content.

Before the workshop

Pre-load materials

Trainer uploads course documents. iQMA ingests them and prepares the knowledge base. Practice exercises are created mapped to the session agenda.

During the workshop

Introduce the platform

Participants are introduced to iQMA as the follow-up layer. Access is granted. The first practice exercise is demonstrated live so it feels familiar before Day 1 ends.

After the workshop

Reinforcement begins

Learners return to practice exercises and quizzes independently. Managers access Team Analytics to see engagement and progress. Capability compounds rather than evaporates.

Team Analytics

Managers and training sponsors get a real-time view of team engagement and progress — which exercises have been attempted, completion rates, and knowledge quiz performance. For the first time, L&D investment has a visible performance signal beyond attendance records.

Manager Visibility

Why It Works

Most AI coaching tools are general-purpose chatbots pointed at a topic. iQMA is different in three specific ways.

Document-grounded, not internet-grounded — the AI coach answers from your specific uploaded materials. It doesn't hallucinate generic answers; it references what was actually taught. Learners trust it because it speaks their language.

Scenario-driven practice, not passive consumption — the coach doesn't present information; it asks questions, diagnoses misunderstandings, and guides the learner to construct their own understanding. This is how adult learning actually works.

Designed by a practitioner, not a software team — iQMA was built by Camil Toorabally, an instructional designer with ACLP and DDDLP credentials from IAL Singapore. The learning architecture reflects evidence-based instructional design principles — Gagné's events of instruction, spaced repetition, and the 70:20:10 model — not product intuition.

Training that ends when the room empties isn't training — it's a very expensive reminder that people need to keep learning. iQMA is what keeps them learning.

Related Case Studies

Maritime & Shipping