NEXURA LABS
← All work

AI + IOT SYSTEM ACADEMIC PROTOTYPE

IntelliClass.

One classroom. Connected systems.

An AI and IoT platform connecting attendance, classroom operations and lecture intelligence. My final-year project at Universiti Malaysia Terengganu.

Explore architecture ↓GitHub ↗
02 / INTELLICLASS
Real IntelliClass student dashboardReal IntelliClass NFC check-in screenReal IntelliClass generated lecture summary
Original Flutter app screens · Academic prototype
Technology track · UMT Student Research Day 2025
2nd place

01

Different users. One system.

IntelliClass — AI-Integrated Smart Classroom for University Learning — brings classroom operations into a central mobile experience for students, lecturers and classroom/facility personnel.

The final-year project at UMT explored attendance validation, lecture processing and utility control together. Requirement discussions with classroom stakeholders helped connect software behavior to the physical environment.

02

Attendance is a workflow, not a tap.

NFC HCE starts the phone-based check-in flow. Attendance stays partially verified until facial verification confirms identity.

The repository contains the mobile face-verification flow, Node.js attendance controller and Python verification modules. The architecture separates initiating a check-in from confirming attendance.

03

Lecture intelligence and IoT control.

Lecture audio processing connects transcription and AI summarization to the app. A separate MQTT path carries classroom utility commands to devices.

Keeping application data and device communication as distinct paths makes the system easier to reason about. The prototype also made hardware availability, API limits and partially functional features visible rather than hiding them.

04

Recognition and perspective.

2nd Place — Best Project Award, Final Year Project II, Technology track, at the 8th Annual UMT Student Research Day 2025.

The project taught me to think across mobile interfaces, APIs, data, identity and physical devices. It remains an academic system, with some simulated or partially functional features; no university-wide rollout is claimed.

COMPONENTS & CONNECTIONS

System architecture.

SYSTEM OVERVIEWConceptual flow →

Application & intelligence

  1. client
    Flutter appStudents, lecturers, technicians
  2. api
    Node.js REST APIsExpress + MySQL
  3. ai
    AI processingTranscription + summaries

Classroom devices

  1. client
    Mobile controlsUtility commands
  2. api
    MQTTDevice messaging
  3. iot
    ESP32 devicesClassroom utilities

Attendance verification

  1. iot
    NFC HCEPhone-based check-in
  2. api
    Partial verificationAwait identity check
  3. ai
    Facial verificationConfirm attendance
IntelliClass — Academic prototype with simulated or partially functional features due to hardware/API limitations. It is not a university-wide deployment.
Owner-supplied IntelliClass architecture connecting Flutter, Node.js Express, MySQL in Docker, Google Cloud Storage and AI APIs, Favoriot, and ESP32 classroom devices

The original project architecture

Original academic design supplied by the project owner. It includes Google Cloud Storage, Speech-to-Text, Gemini, Favoriot and ESP32 subsystems. Design scope does not mean every component was fully implemented.

Open full architecture diagram →

TECHNOLOGY & INTEGRATIONS

Each part has a purpose.

  • Flutter
  • Node.js
  • MySQL
  • Python
  • MQTT
  • ESP32
Flutter / Dart
Role-based mobile application
Google Speech-to-Text
Lecture transcription
Python + generative AI
Facial verification and lecture intelligence
MQTT / ESP32
IoT utility control and attendance hardware

Project scope and implementation references: view the repository ↗. Recognition, professional context and historical metrics are supplied by the project owner.

Continue the conversation.

Interested in the system, the decisions or what comes next?

Get in touch →
NEXT PROJECT / 03Smart Feeder