Company 1 – Tech Flickers

About Tech Flickers

Tech Flickers is the first SECP incubated company of Capital University of Science and Technology (CUST). The company was registered on 25th September, 2020. Tech Flickers is a Software company that helps passionate people and businesses make an impact by providing them different quality services. Our team of experienced developers, graphic designers, content writers, SEO specialist and digital marketers are driven to produce impactful results. We help business like your expands at a rapid scale. We understand the challenges of working with remote team and focus on providing an experience that feels like having a software development team at your door.

Address: 129 lane 4, Scheme 3, Rawalpindi

Key Features
  • Website Development
  • App Development
  • Graphics Designing
  • Content Writing
  • Search Engine Optimization
  • Digital Marketing
  • Programming
Company Website
  • image
  • Pic 1
  • Pic
Company 2 – Vehicle Detection System

About Vehicle Detection System

The Vehicle Detection System is an advanced security solution powered by artificial intelligence, designed to enhance vehicle monitoring and parking management. Utilizing computer vision and machine learning technologies, this system detects vehicles in real-time and identifies their characteristics, such as make, model, and colour. Additionally, it employs optical character recognition (OCR) to read license plates.
Each detected vehicle is verified against a database of pre-registered vehicles, allowing for automated access control and fraud prevention. If a vehicle is registered, it is granted access; if not, entry is denied.

Address: Takhleeq, F-Block, CUST Islamabad

Key Features
  • The admin can register vehicle owners
  • The admin can register vehicles
  • The admin can view registered vehicles
  • The admin can view detection records
  • The admin can view suspicious detections
  • The camera captures images of cars from a video feed
  • Machine learning models extract the vehicle’s make, model, and colour
  • Machine learning models utilize Optical Character Recognition (OCR) to read and extract the vehicle’s number plate
  • The camera application adds the detected information to the database
Unique Features
Traditional System (RFID-Based)
  • Uses RFID stickers placed on vehicle windshields
  • The RFID reader detects the sticker when the vehicle is near the gate
  • System verifies the unique ID from the tag against a database
  • If valid, the barrier gate opens
  • Physical tags can be copied, removed, or misused, reducing security
Improved System (OCR and Vehicle Recognition-Based)
  • No physical tags required, eliminating risks of duplication or misuse
  • The vehicle is detected automatically when it crosses a designated line
  • OCR extracts the number plate from the video feed
  • Vehicle make and colour are classified as an added security layer
  • The extracted details are matched against the database
  • If a match is found, the system grants access and opens the barrier
Technical Description

These are the tools and technologies we are using.

  • Python: The primary programming language for developing the backend services, including the implementation of machine learning models for number plate detection and recognition.
  • PyTorch: We will use these libraries to build and train machine learning models. We applied YOLO and transfer learning to it. YOLO v11 Nano was pretrained on ImageNet.
  • ReactJS & NextJS: The front end of the application will be built using NextJS.
  • NodeJS: was used to write APIs that connect the frontend with the database
  • TailwindCSS: A utility-first CSS framework that allows for rapid UI development with customizable styling, enabling a responsive and modern design
  • PostgreSQL: The system will use PostgreSQL as its database for storing vehicle registration data and timestamps
Host URL
  • 1
  • 3
  • 4
  • 5
  • 2
  • 6
Company 3 – CUST App

About CUST App

The CUST App is a university automation system designed to digitize and simplify core administrative tasks through a mobile-first experience. It empowers students to request official letters, allows external guests to schedule campus visits, and manages student registration for convocations — all from a single platform. Each module includes approval workflows, dynamic PDF generation with auto-filled templates, and real-time notifications. The system supports role-based access for students, staff, security, and admins. Admins can also create reusable templates using placeholders, ensuring consistency and flexibility in communication. The app enhances operational efficiency, reduces manual workload, and provides transparency throughout the process.

Address: Takhleeq, F-Block, CUST Islamabad

Features
  • Mobile-based letter request and approval system
  • Auto-populated official letters with university letterhead and digital signature
  • Guest visit request and approval system with real-time notifications
  • Security portal for scanning CNICs and managing guest check-ins
  • Convocation registration with eligibility upload and invitation generation
  • Dynamic PDF invitation and letter generation system
  • Admin dashboard for managing templates and events
  • Role-based access control for departments, students, security, and admins
  • Status tracking and expiry logic for unfulfilled guest requests
  • Email dispatch system for invitation and letter delivery
Unique Features
  • Dynamic template engine with placeholder-based auto-population
  • Integrated security check-in module with CNIC scanning for guest management
  • Multi-role access with separate interfaces for students, staff, security, and admins
  • Convocation automation with registration, eligibility checks, and smart emailing
  • End-to-end digitization of manual university processes via a single app
  • Automatic request expiry system for no-show guest visits
Technical Description

Built on Django REST Framework with PostgreSQL database and deployed on cloud infrastructure for high availability. Frontend utilizes Bootstrap 4, jQuery, and Material Design with CKEditor for rich text editing. Security implements JWT authentication, CSRF protection, and role-based access control. File processing includes Excel parsing, ReportLab PDF generation, and cloud storage integration. The mobile component is a native Android application with API integration. Technology stack encompasses Python 3.x, Django 4.x, HTML5/CSS3, JavaScript ES6+.

Host URL
  • Screenshot_2025-08-08-12-40-25-26_4341304f829ecdf69236f7a38641c5b7
  • login
  • Screenshot_2025-07-11-21-14-27-07_c60db80ea4fe39653ab1c89891e67cc5
  • Screenshot_2025-08-08-12-07-08-26_4341304f829ecdf69236f7a38641c5b7
  • Screenshot_2025-07-11-21-12-27-03_c60db80ea4fe39653ab1c89891e67cc5
  • homepage
  • employee_requests
  • create_application
Company 4 – FlexTracker

About FlexTracker

FlexTracker is a startup founded by BS Computer Science graduates from the Capital University of Science and Technology, Islamabad. Flex Tracker is a fleet management application that can be employed by different companies/organizations to track their vehicles or goods. This flexible platform manages multiple services in one platform. The system is capable of dynamic stop and route configurations, comparison of followed route by driver with actual defined route, live location of vehicles on map, and over speeding notification to admin. The potential clients for flex tracker could be universities, schools, and goods distributors.

Address: Takhleeq, F-Block, CUST Islamabad

Key Features
  • Real Time Vehicle Tracking
  • Time Saving
  • Reduction in Fuel Cost
  • Driver Management

The company has now dissolved.