Page 1 of 45 - Cover Page Submission ID trn:oid:::27535:97857111

02report_f_1.pdf
Delhi Technological University

Document Details

Submission ID

trn:oid:::27535:97857111 40 Pages

Submission Date 6,606 Words

May 26, 2025, 10:44 PM GMT+5:30

38,996 Characters

Download Date

May 26, 2025, 10:56 PM GMT+5:30

File Name

02report_f_1.pdf

File Size

10.7 MB

Page 1 of 45 - Cover Page Submission ID trn:oid:::27535:97857111

 Page 2 of 45 - Integrity Overview Submission ID trn:oid:::27535:97857111

11% Overall Similarity

The combined total of all matches, including overlapping sources, for each database.

Match Groups Top Sources

54 Not Cited or Quoted 11% 5% Internet sources

Matches with neither in-text citation nor quotation marks
1% Publications
0 Missing Quotations 0% 11% Submitted works (Student Papers)
Matches that are still very similar to source material

0 Missing Citation 0%
Matches that have quotation marks, but no in-text citation

0 Cited and Quoted 0%

Matches with in-text citation present, but no quotation marks

Integrity Flags
0 Integrity Flags for Review
Our system's algorithms look deeply at a document for any inconsistencies that
No suspicious text manipulations found. would set it apart from a normal submission. If we notice something strange, we flag
it for you to review.

A Flag is not necessarily an indicator of a problem. However, we'd recommend you

focus your attention there for further review.

Page 2 of 45 - Integrity Overview Submission ID trn:oid:::27535:97857111

 Page 3 of 45 - Integrity Overview Submission ID trn:oid:::27535:97857111

Match Groups Top Sources

54 Not Cited or Quoted 11% 5% Internet sources

Matches with neither in-text citation nor quotation marks
1% Publications
0 Missing Quotations 0% 11% Submitted works (Student Papers)
Matches that are still very similar to source material

0 Missing Citation 0%
Matches that have quotation marks, but no in-text citation

0 Cited and Quoted 0%

Matches with in-text citation present, but no quotation marks

Top Sources
The sources with the highest number of matches within the submission. Overlapping sources will not be displayed.

1 Internet

dspace.dtu.ac.in:8080 1%

2 Submitted works

Delhi Technological University on 2019-05-30 <1%

3 Internet

www.coursehero.com <1%

4 Submitted works

Royal Thimphu College on 2025-05-25 <1%

5 Submitted works

University of Wales Institute, Cardiff on 2025-05-25 <1%

6 Submitted works

Delhi Technological University on 2018-05-12 <1%

7 Submitted works

University of Greenwich on 2024-12-02 <1%

8 Submitted works

University of Essex on 2025-04-25 <1%

9 Internet

www.dspace.dtu.ac.in:8080 <1%

10 Submitted works

Visvesvaraya Technological University on 2014-11-06 <1%

Page 3 of 45 - Integrity Overview Submission ID trn:oid:::27535:97857111

 Page 4 of 45 - Integrity Overview Submission ID trn:oid:::27535:97857111

11 Submitted works

Liverpool John Moores University on 2024-05-10 <1%

12 Submitted works

University of Birmingham on 2025-05-02 <1%

13 Submitted works

University of Western Australia on 2025-04-06 <1%

14 Submitted works

CITY College, Affiliated Institute of the University of Sheffield on 2021-09-10 <1%

15 Submitted works

IIMT University on 2024-07-08 <1%

16 Submitted works

Liverpool John Moores University on 2025-02-24 <1%

17 Submitted works

Akdeniz University on 2024-09-01 <1%

18 Submitted works

Higher Education Commission Pakistan on 2019-12-06 <1%

19 Internet

pergamos.lib.uoa.gr <1%

20 Submitted works

Universidad Continental on 2024-05-02 <1%

21 Internet

eoscfuture.eu <1%

22 Submitted works

Universidade de Brasília, UnB on 2023-03-02 <1%

23 Submitted works

University of Ulster on 2025-05-09 <1%

24 Submitted works

University of Wales Institute, Cardiff on 2024-08-07 <1%

Page 4 of 45 - Integrity Overview Submission ID trn:oid:::27535:97857111

 Page 5 of 45 - Integrity Overview Submission ID trn:oid:::27535:97857111

25 Submitted works

Loughborough University on 2022-05-03 <1%

26 Submitted works

UCL on 2025-05-02 <1%

27 Internet

ece.anits.edu.in <1%

28 Submitted works

Liverpool John Moores University on 2024-11-18 <1%

29 Submitted works

Swinburne University of Technology on 2024-11-03 <1%

30 Submitted works

University of Leicester LTI on 2025-05-01 <1%

31 Submitted works

University of Westminster on 2025-04-12 <1%

32 Submitted works

ESoft Metro Campus, Sri Lanka on 2025-04-23 <1%

33 Internet

edoc.pub <1%

34 Internet

icandecide.org <1%

35 Submitted works

Delhi Technological University on 2024-05-01 <1%

36 Submitted works

University of Leicester on 2025-05-16 <1%

Page 5 of 45 - Integrity Overview Submission ID trn:oid:::27535:97857111

 Page 6 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

i
Encorp AI
( AI-powered slide generator)

1 A PROJECT REPORT
SUBMITTED IN PARTIAL FULFILLMENT OFTHE
REQUIREMENTS FOR THE AWARD OF DEGREE
OF
BACHELOR OF TECHNOLOGY
IN
SOFTWARE ENGINEERING

Submitted By
Md Rizabul
4 (2K22/SE/114)
Shahrukh Khan
(2K22/SE/160)

Under the supervision of

MS. PRIYA SINGH
ASSITANT PROFESSOR

4 DEPARTMENT OF SOFTWARE ENGINEERING

DELHI TECHNOLOGICAL UNIVERSITY

(Formerly Delhi College of Engineering)

Bawana Road, Delhi – 110042

Page 6 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 7 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

ii

DEPARTMENT OF SOFTWARE ENGINEERING

DELHI TECHNOLOGICAL UNIVERSITY
(Formerly Delhi College of Engineering)
Bawana Road, Delhi - 110042

CANDIDATE’S DECLARATION

We, Md Rizabul (2K22/SE/114) and Shahrukh Khan (2K22/SE/160), students of B.Tech (Software

3 Engineering) declare that the project report titled “Encorp-AI(AI-powered slide generator)” which

is submitted by us to the Department of Software Engineering, Delhi Technological University,

Delhi in partial fulfillment of the requirement for the award of the degree of Bachelor of

Technology, is original and not copied from any source without any proper citation. This work has

not previously formed the basis for the award of any Degree, Diploma Associateship, Fellowship or

1 other similar title or recognition. The work has been accepted in peer reviewed Scopus indexed

conference with the following details:

Place: Delhi MD RIZABUL SHAHRUKH KHAN

Date (2K22/SE/114) (2K22/SE/160)

Page 7 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 8 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

9 iii

DEPARTMENT OF SOFTWARE ENGINEERING

DELHI TECHNOLOGICAL UNIVERSITY
(Formerly Delhi College of Engineering)
Bawana Road, Delhi - 110042

CERTIFICATE

35 We hereby certify that the project titled “Encorp-A1(AI-powered slide generator)”, which is

2 submitted by Shahrukh Khan, Roll No: 2K22/SE/160 and Md Rizabul, Roll No: 2K22/SE/114,

Department of Software Engineering, Delhi Technological University, Delhi in partial fulfillment of

the requirement for the award of the degree of Bachelor of Technology, is a record of project work

carried out by students under my supervision. To the best of my knowledge, this work has not been

submitted in part or full for any Degree or Diploma to this University or elsewhere.

Place: Delhi Ms. PRIYA SINGH

Date: SUPERVISOR
(Assistant Professor, SE , DTU)

Page 8 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 9 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

iv
6
ACKNOWLEDGEMENT

We would like to express our gratitude and appreciation to all those who gave us the opportunity to

complete this dissertation. Special thanks to our supervisor in charge, Ms. Priya Singh, Department

6 of Software Engineering whose help, stimulating suggestions and encouragement helped and guided

us for the completion of the project undertaken by me. It is with their supervision that this work

1 came into existence. We would like to thank the department of Software Engineering for providing

the infrastructure, facilities and opportunity to work in this knowledgeable project that helped me to

learn new things about this burning technology. We also want to give our special thanks to all our

fellow mates who supported us in every course and provided their valuable ideas and thoughts to us.

SHAHRUKH KHAN MD RIZABUL

2K22/SE/160 2K22/SE/114

Page 9 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 10 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

V
ABSTRACT

In the age of rapid digitization and content generation, the demand for high-quality, visually

appealing, and contextually accurate presentations has surged across industries, academia, and

enterprises. However, creating professional slide decks manually can be time-consuming, repetitive,

and creatively taxing for users, especially when working under tight deadlines. To address this gap,

Encorp presents an AI-powered solution that automates the generation of complete presentation

slides based on minimal user input.

This project leverages the power of Next.js for front-end development, Supabase for authentication

and data storage, and Gemini API to generate human-like slide content. Users can simply enter a

topic, select the desired number of slides, and choose a presentation style to receive a ready-to-

download .pptx file. Features such as user authentication, history tracking, preview before export,

and theming have been seamlessly integrated using React, TailwindCSS, NextAuth, and

PPTXGenJS.

Encorp not only saves time and effort but also ensures consistency, clarity, and customization,

making it an ideal tool for students, professionals, and educators. The application demonstrates the

potential of large language models in practical real-world applications, and paves the way for future

enhancements such as voice input, multilingual support, and deeper integration with cloud

collaboration tools.

Page 10 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 11 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

vi

CONTENTS

TABLE OF CONTENTS PAGE NO

TITLE PAGE I i
1 CANDIDATE’S DECLARATION ii
CERTIFICATE iii
ACKNOWLEDGEMENT iv
ABSTRACT v
CONTENTS vii
1 CHAPTER 1 INTRODUCTION 1
1.1 Background and Motivation 1
1.2 Problem Statement 2
1.3 Signi cance of the project 3
CHAPTER 2 OBJECTIVES 4
2.1 Primary Objectives 4
34 2.2 Technical Objectives 4
2.3 Educational Objectives 4
2.4 User-Centric Objectives 5
2.5 Validation Objectives 5
2.6 Future-Oriented Objectives 5
CHAPTER 3. SYSTEM ARCHITECTURE 6
3.1 User Interface 6
3.2 Backend Logic 7
3.3 Redis Queue and Worker Architecture 8
18 3.4 Export and File Generation Module 9
3.5 Deployment Architecture 10
3.6 Architectural Advantages 11
3.7 Database Layer 11
3.8 Scalability and Reliability 11
CHAPTER 4 METHEDOLOGY 12
4.1 Frontend 12
4.2 Backened 13
4.3 Database 14
4.4 API integration 15
CHAPTER 5 FEATURES 16
5.1 Authentication System 16
5.2 Dashboard 16

Page 11 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

fi
 Page 12 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

5.3 Create Presentation 16

5.4 Real-Time Job Handling 17
5.5 AI-Generated Slides 17
5.6 Responsive Design and Accessibility 18
5.7 User Experience Features 18
5.8. Security and Privacy 18
CHATPTER 6 TECHNOLOGY STACK 19
20 6.1 Frontend Technologies 19
6.2 Backend Technologies 19
6.3 Database Technologies 20
6.4 APIs and AI Integration 20
6.5 State Management and Data Fetching 21
6.6 Job Queue and Background Processing 21
6.7 Deployment and Analytics 21
6.8 Development and Tooling 22
10 CHAPTER 7. IMPLEMENTATION DETAILS. 23
7.1 Authentication (Login & Signup) 23
7.2 Dashboard Implementation 23
7.3 Slide Generation (Create Page) 23
7.4 AI Integration and Job Processing 24
7.5 Slide Preview and Download 24
7.6 Responsive and Themed UI. 25
7.7 Noti cations and Feedback. 25
7.8 Error Handling and Edge Cases. 25
7.9 Deployment and Hosting. 25
7.10 Final Touches. 25
CHAPTER 8. RESULTS 26
8.1 Landing page of Encorp-AI: 26
8.2 Dashboard 27
8.3 creating form 27
8.4 Progress animation 27
8.5 generated pages 28
8.6 Exporting slides/presentation 29
10 CHAPTER 9. CONCLUSION AND FUTURE SCOPE 30
9.1 Conclusion 30
9.2 Future Scope 30
REFERENCES 32

APPENDICES 33

Page 12 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

fi
 Page 13 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

1
19 CHAPTER 1

INTRODUCTION

1.1 Background and Motivation

In today’s fast-paced world, creating professional slide decks quickly and effectively is more
important than ever — whether it's for pitches, classrooms, or internal briefings. However, most
people either struggle with structuring their thoughts into clear slides or end up spending too much
time formatting presentations rather than focusing on content. That’s where this project — an AI-
powered slide generation platform — comes in.

The core idea is simple: what if you could type a prompt like “Create a presentation about the
impact of AI in education”, and instantly receive a structured slide deck — complete with bullet
points, descriptions, and ready-to-export formats like PDF or PPT? That’s the experience we aimed
to deliver by combining the power of Google’s Gemini AI with a fast, scalable, and modular
backend architecture.

But beyond the user experience, this project was also an opportunity to explore how real-time,
event-driven systems can be built in a resource-efficient way. Instead of overloading the server with
constant polling or making the AI process run synchronously, we built a Redis-backed job queue
that triggers Golang worker services the moment a task arrives. This makes the system fast, cost-
effective, and highly scalable — especially useful when more users and tasks are introduced.

The motivation for building this project was both practical and academic. On one hand, we wanted
to simplify a task that many people do daily — making slides. On the other, we were curious to
explore how modern backend patterns (job queues, workers, AI integrations) can be built and
deployed efficiently using tools like Redis, EC2, and cloud APIs.

Looking ahead, there are many directions for growth: adding cloud uploads to S3 or Cloudinary,
enabling image generation for richer visuals, and expanding the use of AI to include things like
voice-to-slide generation or multilingual presentation support.

In short, this project represents a blend of real-world utility and deep backend engineering, all
centered around making everyday workflows smarter with AI.

Page 13 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 14 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

1.2 Problem Statement

Creating structured, visually appealing, and informative slide presentations is a time-consuming

task that often requires both domain knowledge and design skills. Whether for business, education,
or research, users spend signi cant time drafting, organizing, and formatting slides — time that
could be better spent re ning ideas or preparing for delivery.

Despite the rise of generative AI, there is still a gap in tools that allow users to go from a raw idea
or prompt to a fully usable presentation with minimal manual effort. Existing AI-based tools either
lack customizability, suffer from slow processing, or do not provide export options in standard
formats like PDF or PPT.

Additionally, many AI-integrated applications suffer from inef cient backend designs, leading to
high latency, resource wastage due to constant polling, and poor scalability when multiple users
interact with the system simultaneously.

The key challenges this project aims to address are:

• How to convert natural language prompts into structured, slide-ready content.

• How to process AI tasks asynchronously with real-time responsiveness.

• How to maintain a scalable, ef cient, and resource-conscious backend.

• How to provide users with exportable and editable formats of the generated output.

This project proposes a solution that combines Gemini AI, Redis-based job queueing, and a
microservice-based architecture to provide a fast, reliable, and user-friendly slide generation tool.
The goal is to bridge the gap between idea generation and professional presentation — all within a
few clicks.

Page 14 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

fi
fi
fi
fi
 Page 15 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

33 1.3 Signi cance of the project

This project holds signi cance both in practical application and in the technical approach it adopts.
In an age where speed, clarity, and automation are crucial, the ability to instantly convert ideas into
structured slide presentations can save valuable time and effort for students, educators,
professionals, and businesses alike.

From a user perspective, this platform transforms a tedious and repetitive process into a seamless
experience. By simply entering a prompt, users receive AI-generated, ready-to-export slides —
reducing the time required for content research, slide structuring, and formatting.

From a technical standpoint, the project demonstrates a robust and scalable architecture that
embodies real-world software engineering practices. It combines multiple cutting-edge technologies
— including:

• Generative AI (Gemini) for content creation,

• Redis Pub/Sub and Queueing for ef cient background processing,

• Golang workers for asynchronous task execution,

• Cloud deployment (Vercel + AWS EC2) for scalability and availability,

• And export options (PDF and PPT) to meet practical user needs.

This makes the project a valuable reference for building AI-integrated systems with real-time
capabilities and resource-ef cient designs.

Moreover, the platform contributes to a growing interest in AI-as-a-service, showing how large
language models can be practically applied beyond chat interfaces — into structured document
generation, automation tools, and knowledge delivery systems.

By bridging user-friendliness with technical sophistication, this project not only solves a common
problem but also serves as a blueprint for future AI-driven productivity tools. Its modular and
scalable architecture opens doors for further innovations such as image integration, voice-based
input, team collaboration features, and enterprise-grade deployment.

In summary, this project is signi cant for:

• Enhancing productivity through automation,

• Demonstrating modern backend design patterns,

• Bringing AI closer to end-users through a usable, deployable interface,

• And inspiring future work in AI-based content generation tools.

Page 15 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

fi
fi
fi
fi
fi
 Page 16 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

4
CHAPTER 2

OBJECTIVES

This chapter outlines the various goals this project aims to achieve, categorized into primary,
technical, educational, user-centric, validation, and future-oriented objectives.

2.1 Primary Objectives

• To develop a web-based platform that generates structured slide presentations from natural
language prompts using AI.

• To automate the process of slide creation, minimizing the manual effort needed for research,
structuring, and formatting.

• To deliver a complete end-to-end user experience with content generation, real-time

processing, previewing, and exporting.

2.2 Technical Objectives

• To integrate Gemini AI for processing and generating slide content from user prompts.

• To design and implement a Redis-backed job queue system for asynchronous task
management.

• To develop a Golang-based worker that consumes tasks and interacts with the Gemini API.

• To build a secure and scalable Node.js backend for authentication, API handling, and
database interactions.

• To implement PDF and PPT export functionalities using jsPDF and pptgenx.

• To deploy the application efficiently using Vercel (frontend) and AWS EC2 (backend and
worker services).

2.3 Educational Objectives

• To gain hands-on experience with microservices architecture and distributed task

processing.

• To understand the integration of AI models into full-stack applications.

• To explore and apply advanced concepts in Redis, Pub/Sub systems, and queue-based
workflows.

• To enhance proficiency in using cloud platforms (AWS, Vercel) and DevOps practices
(environment setup, deployment, scalability).

Page 16 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 17 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

5
2.4 User-Centric Objectives

• To provide users with a simple, intuitive interface to generate and preview slide content.

• To ensure real-time responsiveness by avoiding long wait times or UI freezes during AI

processing.

• To enable export options (PDF and PPT) that match user expectations for convenience and
usability.

• To offer a consistent and bug-free experience across different devices and browsers.

2.5 Validation Objectives

• To validate the correctness and coherence of slide content generated by the AI.

• To evaluate system performance in terms of latency, error handling, and queue reliability.

• To test the scalability of the system under varying loads by simulating multiple user
prompts.

• To verify successful integration of all system components — frontend, backend, AI

processing, and export utilities.

2.6 Future-Oriented Objectives

• To implement cloud upload functionality using platforms like S3 or Cloudinary for

persistent storage of generated files.

• To enhance PDF generation by including relevant images, once affordable image

generation APIs are integrated.

• To support voice-to-prompt input and multilingual generation in future versions.

• To evolve the platform into a broader AI productivity tool, expanding beyond slide
generation.

Page 17 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 18 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

6
CHAPTER 3

SYSTEM ARCHITECTURE

The system follows a modular, event-driven architecture, combining a modern web frontend with a
7 robust backend and asynchronous worker services. It is designed for responsiveness, scalability, and
efficient integration with AI services. This chapter elaborates on the roles and technologies of each
component involved in the system.

21 3.1 User Interface

The user interface is the primary point of interaction for end-users, allowing them to input prompts,
preview slide content, and export the results. It emphasizes simplicity and responsiveness to ensure
a smooth user experience.

Key Features of the User Interface:

1.Clean and modern slide creation interface

2.Form-based input system for prompt submission

3.Real-time status updates for AI job completion

4.Slide preview panel with both bullet and paragraph formats

5.Export buttons for PDF and PPT generation

6.Responsive design with light/dark theme support

Technologies Used:

Framework: Next.js (v14.1.3) – React-based framework for server-side rendering and routing

Styling: TailwindCSS, Tailwind Merge, Tailwind Animate

Component Libraries:

• Radix UI: Multiple packages for accessible and customizable components (modals,
dropdowns, sliders, dialogs)

• Framer Motion: For smooth animations

• Lucide-react: Icon library

State & Form Management:

36 • react-hook-form with zod for schema validation

Page 18 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 19 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

31 • @tanstack/react-query for asynchronous data fetching and caching

Data Visualization: Recharts for potential analytics views

Export Tools:

• jsPDF for PDF generation

• pptxgenjs for PowerPoint export

Authentication: next-auth for session-based login

Analytics: @vercel/analytics

Others: axios, clsx, date-fns, html2canvas, cmdk, sonner (toasts), etc.

3.2 Backend Logic

The backend acts as the central control system, handling client requests, authentication, and
communication with both the database and Redis-based queueing system.

Key Responsibilities:

• Handles user authentication and token generation

• Accepts user prompts and creates a job entry in Redis

• Publishes job notifications to the Redis Pub/Sub channel

• Provides APIs to poll job status

• Persists job data to PostgreSQL for logging/tracking

Technologies Used:

• Runtime: Node.js (v18+)

• Framework: Express.js

• Authentication: JWT-based session validation with secure secrets

• Database: PostgreSQL (via Supabase or containerized instance)

• Queueing System: Redis (via Upstash or local deployment)

• Environment Config: Managed through .env files

Page 19 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 20 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

Key Responsibilities:

• Acts as the bridge between frontend requests and backend APIs

• Transforms user input into a task format

• Receives task status and results

• Updates the frontend upon job completion

Technologies Used:

• API Communication: RESTful APIs via axios

• Client-side Cache & Sync: react-query

Third-Party Services Connected:

• Gemini AI API: Processes prompts and returns structured slide data

• PostgreSQL: Stores session and metadata

• Redis: Manages task queue and Pub/Sub for event-driven job triggering

• Vercel Analytics: Tracks usage patterns

• AWS EC2

3.3 Redis Queue and Worker Architecture

The core innovation of this system lies in its asynchronous and event-driven background
processing. Tasks are decoupled from the main user request, enabling faster responses and high
throughput.

Key Responsibilities:

• Listens to Redis Pub/Sub noti cations

• Fetches queued jobs from Redis

• Sends prompts to Gemini API

• Receives structured slide content (JSON)

• Updates Redis with job results and status

Technologies Used:

• Language: Golang (v1.20+)

Page 20 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

fi
 Page 21 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

• Redis Client: Go Redis library

• HTTP Client: Go standard net/http

• Gemini AI Integration: Authenticated using API key

• Concurrency Model: Goroutines for efficient processing

Workflow Summary:

1. User submits a prompt.

2. Backend pushes the task to a Redis queue.

3. Redis Pub/Sub notifies the Golang worker instantly.

4. Worker fetches the task, sends the prompt to Gemini API.

5. Once processed, the result is pushed back to Redis and job status is updated.

6. Frontend polls backend until job status is marked "completed", then displays the slides

Figure A1
3.4 Export and File Generation Module

This module handles the conversion of structured slide content into downloadable formats such as
PDF and PPT.

Responsibilities:

• Converts slide data into exportable formats

• Supports:

Page 21 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 22 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

10

◦ PDF Export using jsPDF

◦ PPT Export using pptxgenjs

• Prepares files for future upload to S3 or Cloudinary

Future Enhancements:

• Support for image generation

• Cloud uploads

• Export history and versioning

3.5 Deployment Architecture

A cloud-based deployment approach ensures high availability and performance for the system
components.

Components & Platforms:

• Frontend: Deployed on Vercel for fast CDN-backed delivery

• Backend & Golang Worker: Deployed on AWS EC2, managed via environment variables

• Redis & PostgreSQL: Can be self-hosted or obtained from Upstash, Aiven, or Supabase

Page 22 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 23 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

11

3.6 Architectural Advantages

• Real-Time Response: Pub/Sub model ensures low latency

• Scalability: Horizontal scaling by adding multiple Golang workers

• Separation of Concerns: Distinct services for UI, backend, worker

• Resource Efficiency: Workers remain idle until jobs arrive

• Robustness: Redis ensures no task is lost, even under failure

3.7 Database Layer

Responsibilities:

• Persist user-related metadata, job details, and prompt/response pairs.

• Store authentication sessions (when using Supabase Auth or NextAuth.js with a database
adapter).

• Log interaction history for future audit, debugging, or analytics.

• Serve as a reliable storage point in case Redis data is lost after TTL expiry.

Database Schema Overview (Simplified):

3.8 Scalability and Reliability

• Stateless backend and workers allow horizontal scaling.

• Redis Pub/Sub enables low-latency real-time processing without polling overhead.

• Workers are designed to auto-recover from task failures to ensure reliability.

Page 23 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 24 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

12
CHAPTER 4

METHEDOLOGY

The development of Encorp-AI followed a modular and scalable approach, combining robust
backend processing, efficient job queuing, and a modern frontend interface. Each component of the
system was developed to fulfill specific roles while maintaining interoperability with others.

4.1 Frontend

The frontend of Syncify is responsible for the user-facing part of the application, where users
12 interact with the system through a visually appealing and functional interface. It was developed
using Next.js, a React-based framework, and styled with Tailwind CSS.

Key Features of the Frontend

1. Dynamic and Responsive Design:

12 ◦ Built with Tailwind CSS to ensure a responsive layout across various devices.
◦ Components adapt seamlessly to mobile, tablet, and desktop resolutions.
16 2. Server-Side Rendering (SSR) and Static Site Generation (SSG):

◦ Next.js's SSR ensures fast page loads and better SEO by pre-rendering pages on the
server.
◦ Pages requiring real-time updates leverage Client-Side Rendering (CSR).
3. User-Friendly Workflow Management:

◦ Interactive panels allow users to configure workflows with minimal effort.

◦ Intuitive drag-and-drop components simplify workflow creation.
4. Real-Time Feedback and Updates:

◦ Use of Next.js API routes to fetch data dynamically.

◦ Notifications and progress bars provide immediate feedback for user actions.

Page 24 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 25 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

13

Technology Stack:

• Framework: Next.js (React-based)

• Styling: TailwindCSS with Radix UI components

• State Management: React Query for API data caching and mutation

• Animations: Framer Motion

• Form Handling & Validation: React Hook Form + Zod

Approach:

• Developed reusable components for slide creation, prompt input, and job status updates.

• Implemented dynamic routes and API integration using Next.js app and pages directory
structure.

• Used tailwindcss-animate and motion libraries to enhance interactivity.

• Integrated jsPDF and pptxgenjs for download/export features.

• Designed for responsiveness and accessibility with mobile-friendly layouts.

4.2 Backend

The backend of Syncify was developed using Node.js and Express.js, providing the logic and
services necessary to process requests and execute workflows. This layer acts as the system's brain,
28 ensuring smooth communication between the frontend, database, and external APIs.

Key Responsibilities:

1. User Authentication and Authorization: Handles secure user login and access control
using JWT (JSON Web Tokens).
2. Workflow Processing: Manages the creation, modification, and execution of user-defined
automation workflows.
3. API Communication: Facilitates interactions with third-party services (e.g., Slack, Twitter)
via RESTful APIs.
4. Error Handling: Logs errors and provides appropriate responses to the frontend.

Page 25 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 26 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

14

Technology Stack:

• Language: Node.js (Express.js)

• Authentication: NextAuth.js or Supabase Auth

• Queue Integration: Redis for job queueing and Pub/Sub

• API Security: JWT-based authentication

• Task Routing: RESTful API routes

Approach:

• Designed RESTful APIs for login, prompt submission, and job tracking.

• Authenticated users with token-based access control.

• Connected to Redis to enqueue tasks and publish events.

• Used environment-driven configuration for flexible deployment (local/dev/prod).

• Validated API requests using Zod schemas to ensure strict input/output structure.

4.3 Databse

Technology Stack:

• Database: PostgreSQL

• Management Tool: Supabase (hosted) or Docker PostgreSQL (self-hosted)

• Schema Definition: SQL + Supabase CLI

• ORM (optional): Supabase Client / Prisma

Approach:

• Structured relational schema with normalized tables: users, jobs, responses, sessions.

• Stored job metadata and user interactions for traceability.

• Utilized Supabase’s real-time and Auth features for simplified user/session management.

• Maintained database integrity using foreign keys and indexed queries.

Database Schema Overview:

Page 26 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 27 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

15

24 1. Users Table: Stores user information such as id, name, email, password_hash.
2. Work ows Table: Tracks work ows with details like id, user_id, trigger,
actions, and status.
3. Logs Table: Records logs with elds such as id, workflow_id, timestamp, and
execution_result.

4.4 API integration

Key APIs and Endpoints:

• POST /api/submit: Accepts user prompt and queues task

• GET /api/status/:jobId: Checks job status

• GET /api/result/:jobId: Fetches generated slide JSON

• POST /api/login: User authentication

• GET /api/logout: End user session

Integration Details:

• Connected frontend via Axios to these backend routes.

• Used HTTP status codes to indicate job lifecycle (202 Accepted, 200 OK, 500 Error).

• Polling mechanism implemented on frontend to continuously fetch job status.

• Token-based validation middleware for secure API access.

Third-Party APIs:

• Gemini API (Google AI Studio): Used in Golang Worker to process prompt and receive
structured content.

• Redis Pub/Sub API: Used for worker notification and real-time task dispatch.

Page 27 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

fl
fi
fl
 Page 28 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

16
CHAPTER 5

Features

Encorp-AI is a next-generation AI-powered slide generation platform that allows users to transform
textual descriptions into professional presentations with ease. The platform combines seamless UI,
intelligent backend processing, and integration with AI models to deliver accurate and well-
structured presentation slides. Below are the key features of the system:

5.1 Authentication System

5.1.1 User Registration (Sign Up)

25 • New users can create an account securely using an email and password combination.

• Form validation ensures proper input format using react-hook-form and zod.

• Passwords are securely handled via hashing and encrypted storage.

15 5.1.2 User Login

• Existing users can log in to their accounts using secure credentials.

• Session-based authentication implemented using NextAuth or Supabase Auth.

• Invalid attempts are handled gracefully with clear error messages.

5.1.3 Session Persistence

• Authenticated sessions are persisted using cookies or tokens.

• Users remain logged in across sessions until they explicitly sign out.

5.1.4 Logout

• Secure logout mechanism to terminate session and redirect to homepage or login page

• .

5.2 Dashboard

• A personalized landing page after login:

Features:

• Displays a list of previously generated presentations.

• Enables users to preview, download, or recreate slides.

Page 28 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 29 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

17

• Each card shows:

◦ Presentation title or topic

◦ Date and time of creation

◦ Quick-access action buttons (like "Download PDF", "Edit", or "Delete")

•
5.3 Create Presentation

• A powerful input form to generate presentations:

Features:

• Prompt Input: Users can describe the topic in natural language (e.g., "Explain quantum
computing to beginners").

• Number of Slides: Specify how many slides the generated presentation should contain.

• Presentation Style Selector: Choose the tone or aesthetic (e.g., academic, minimalist,
business, creative).

• Submit Button: Initiates background processing via API.

• Loading Animation/Status: Shows task submission success and loading indicator.

5.4 Real-Time Job Handling

• Once the prompt is submitted:

◦ A job is enqueued in Redis and processed by the backend worker.

◦ Users are redirected to a status page while the job is being processed.

Features:

• Polling Mechanism: Frontend periodically checks job status via GET /api/status/:jobId.

• Status Indicators: Show current job progress (e.g., "In Progress", "Generating",
"Completed")

5.5 AI-Generated Slides

• After processing, the AI returns structured content:

Page 29 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 30 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

18

◦ Slide Title

◦ Slide Body (Bullet points, summaries)

◦ Optional notes for the speaker

Output Features:

• Users can preview the result in an interactive UI.

• Can download the result in:

◦ PDF format using jsPDF

◦ PPTX format using pptxgenjs

5.6 Responsive Design and Accessibility

• Fully responsive across devices (mobile, tablet, desktop).

• Utilizes Radix UI components for accessibility compliance (keyboard nav, ARIA tags).

• Clean dark/light themes via next-themes.

5.7 User Experience Features

• Toast Notifications: Inform users of actions like job submission, errors, or download
completion using sonner.

• Progress Indicators: Informative and animated job states.

• Form Auto-validation: Avoids incorrect input submission.

5.8 Security and Privacy

• Secure form submission via HTTPS.

• Server-side validation of prompts and API tokens.

• Role-based access (only logged-in users can generate or view presentations).

Page 30 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 31 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

19

CHAPTER 6

TECHNOLOGY STACK

The Encorp-AI project leverages a robust, scalable, and modern technology stack that ensures high
performance, maintainability, and a seamless user experience. The system is composed of multiple
layers including the frontend, backend, database, APIs, third-party integrations, and deployment
7 tools. Below is a comprehensive breakdown of all the technologies used in the project:

6.1 Frontend Technologies

26 The frontend of Encorp-AI is built using Next.js, a React-based framework optimized for server-
side rendering, routing, and performance. It delivers a fast, interactive, and accessible user interface.

Technology Purpose
7 Next.js (v14+) React framework for SSR, routing, and optimized builds
React (v18.2) Core UI library

TypeScript Strongly-typed JavaScript for safer and scalable development

TailwindCSS (v3.4+) Utility- rst CSS for rapid UI styling

Framer Motion Animation library for smooth transitions and interactive UI
Radix UI Accessible and unstyled components for UI building blocks
Shadcn/UI Reusable Tailwind-compatible component library

clsx / class-variance-authority Conditional class handling for styling

Date-fns Utility library for manipulating dates in JS

Lucide-react Icon library used throughout the UI
17 React-Hook-Form + Zod Form management and schema-based validation
React-Resizable-Panels Dynamic layout management in split views or panels
React-Day-Picker Interactive date selection component
Next-Themes Dark/Light theme toggling
Input-OTP Used for handling OTP-based login ows (if used)

6.2 Backend Technologies

The backend handles business logic, job management, user authentication, and API requests. It uses
Next.js API Routes combined with background workers.

Page 31 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

fi
fl
 Page 32 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

20

Technologies Used:

Technology Purpose

Next.js API Routes API endpoints handled directly in Next.js

Node.js Server-side JavaScript runtime

TypeScript Typed backend logic to reduce bugs

Redis In-memory data store for managing background job queues

Axios Promise-based HTTP client for calling internal or external APIs

Zod Schema-based validation in both frontend and backend

Next-Auth / Supabase Auth User session management and authentication provider

Bcrypt / Supabase For password encryption and login veri cation

6.3 Database

15 Data such as user credentials, job submissions, and history are stored in a reliable and scalable
relational database. Supabase (PostgreSQL-based) was chosen for its seamless integration with
Next.js and real-time capabilities.

Technologies Used:

Technology Purpose

Supabase DB (PostgreSQL) Used as the primary database for storing users, jobs, and metadata

Supabase Auth Integrated user authentication system

Supabase Client SDK To securely access the database from client/backend
Prisma Type-safe ORM for interacting with database (if used)

6.4 APIs and AI Integration

At the core of Encorp-AI lies the integration with large language models to generate slide content.
APIs are used for prompt submission and polling the result.

Page 32 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

fi
 Page 33 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

21
Technologies Used:
Technology / Service Purpose

Gemini APIs Natural language processing to generate structured slide content

REST APIs (internal) For job status, creation, and fetching of results

Axios API client used for HTTP requests between frontend and backend

html2canvas Used to capture UI elements as images (for previews)

jsPDF To export slide content as PDF

To export slide content as downloadable PowerPoint

pptxgenjs
presentations

6.5 State Management and Data Fetching

Efficient state and cache management is essential for smooth user experience and real-time
feedback on jobs.

Technology Purpose
TanStack React Query Caching, refetching, and managing async server state
Next.js ISR / SSR Optimizing data rendering and performance

6.6 Job Queue and Background Processing

To ensure scalability and responsiveness, heavy slide-generation tasks are of oaded to a job queue
system.

Technology Purpose
Redis In-memory job queue storage
BullMQ / Custom Queue Worker Background job processing engine

Polling with React Query To periodically check the status of the user job

6.7 Deployment and Analytics

Encorp is deployed on a production-grade platform to ensure performance and monitoring.

Page 33 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

fl
 Page 34 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

22

Technology Purpose
Vercel Frontend and backend deployment (Next.js optimized platform)
@vercel/analytics Real-time performance and usage insights
Supabase Hosting Manages authentication and database
GitHub Version control and collaboration
CI/CD via Vercel Automatic deployment on push to main branches

6.8 Development and Tooling

Ef cient development is ensured using best-in-class tools

Tool/Library Purpose

ESLint + eslint-con g-next Code quality and linting standards

Prettier (optional) Code formatting

PostCSS + Autopre xer CSS processing and vendor pre xing

VSCode Code editor

Chrome DevTools Frontend debugging and performance pro ling

Page 34 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

fi
fi
fi
fi
fi
 Page 35 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

23

CHAPTER 7

IMPLEMENTATION

The development of Encorp-AI was done with a user-first approach, aiming to make the slide
creation process as seamless and intelligent as possible. This chapter provides an overview of how
different components of the application were implemented, starting from authentication to slide
generation, all the way to downloading final presentations.

7.1 Authentication (Login & Signup)

The project uses Supabase Auth to manage user authentication. It supports traditional email and
password login. Passwords are securely hashed and stored, and sessions are managed using secure
JWT tokens. Users must log in to access features like creating slides, viewing history, or managing
their dashboard.

• Signup includes validation using React-Hook-Form and Zod to ensure data integrity.

• On successful signup or login, a session is created and stored in cookies using Next-Auth or
Supabase's built-in session manager.

• Protected routes ensure users must be authenticated to access dashboard and create pages

7.2 Dashboard Implementation

Once logged in, users land on the dashboard. This page acts as a central hub for tracking previous
slide generation requests.

• The frontend fetches the user's job history using React Query to ensure smooth data
fetching and caching.

• Each job entry displays its status (e.g., pending, generating, completed) and has options to
preview, download, or regenerate slides.

• If no jobs are present, the user sees a friendly message encouraging them to create their first
slide

7.3 Slide Generation (Create Page)

29 The /create page is the heart of the application. It allows users to input the required details for their
AI-generated presentation.

Features include:

• A text input for description or topic of the presentation.

Page 35 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 36 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

24

• A dropdown to select the presentation style (e.g., minimalist, business, educational).

• A number input to define how many slides the user wants.

• All inputs are validated in real-time using Zod and React Hook Form.

Once the form is submitted:

• A POST request is sent to the backend API route.

• This creates a background job and stores its metadata in the Supabase database.

• A loading animation is shown while the job is processed.

7.4 AI Integration and Job Processing

After form submission, the backend triggers an AI generation process:

• The backend formats a prompt and sends it to the OpenAI GPT API (either GPT-3.5 or
GPT-4).

• The API responds with structured slide content broken down into a title, bullet points, and
optionally speaker notes.

• This raw response is cleaned and saved as part of a user’s job.

To avoid blocking the main thread:

• Jobs are queued in Redis and processed asynchronously.

• The frontend polls the backend every few seconds using React Query’s polling feature to
check if the job is complete.

7.5 Slide Preview and Download

Once the job is marked as “completed”:

• The user is redirected to a preview page.

• Here, slides are shown one by one, styled with TailwindCSS and responsive to different
screen sizes.

• Users can download the slides as:

◦ PDF (generated using jsPDF)

◦ PPTX (via pptxgenjs library)

This adds flexibility depending on how the user wants to present or edit the slides.

Page 36 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 37 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

25

7.6 Responsive and Themed UI

The entire interface is fully responsive and supports both light and dark themes.

• Theme switching is handled by Next-Themes, with user preferences saved in local storage.

• TailwindCSS ensures mobile-first styling and accessibility.

• UI components use Shadcn/UI and Radix UI to offer a polished and interactive experience.

7.7 Notifications and Feedback

User feedback is critical. Therefore:

• Sonner is used for showing toast notifications — success on job creation, errors, and other
messages.

• Edge cases like empty inputs, API failure, or job timeout are gracefully handled with
appropriate alerts.

7.8 Error Handling and Edge Cases

• All API routes have error handling using try-catch blocks.

• Validation is implemented both on the client and server using Zod.

• If a job fails or the AI returns an incomplete response, the system retries once or shows an
error notification.

11 7.9 Deployment and Hosting

• The application is hosted on Vercel, which ensures fast and reliable global delivery.

• Backend APIs and server-side rendering are handled by Next.js’s built-in features.

• The database and authentication services are powered by Supabase, hosted and managed
with real-time updates.

7.10 Final Touches

• Icons are powered by Lucide-react, giving a clean and modern look.

• Forms, dropdowns, and modals are enhanced using Radix UI primitives.The loading and
transition animations are added using Framer Motion to keep the user engaged during
interactions.

Page 37 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 38 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

26
CHAPTER 8

RESULT

8.1 Landing page of Encorp-AI:

Figure B1

Page 38 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 39 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

27
8.2 Dashboard:

Dashboard includes all the previous PPT completed by the user and provides button for creating new one

Figure B2

8.3 creating form

It asks Descriptions ,Style and Options for number of slides - 5, 10 ,15, 20

Figure B3
8.4 progress animation

Page 39 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 40 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

28

Figure B4

8.5 generated pages

Here user can preview before any export, using next and previous buttons

Figure B5

Page 40 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 41 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

29
8.6 Exporting slides/presentation

Can be done by using export button in top right corner either in PPT or PDF form

Figure B6

Figure B7

Page 41 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 42 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

30
27 CHAPTER 9

CONCLUSION AND FUTURE SCOPE

9.1 Conclusion

The development of Encorp-AI has successfully demonstrated how arti cial intelligence can
streamline the process of creating high-quality, visually appealing presentations. By integrating
30 OpenAI's language models with modern web technologies such as Next.js, Supabase,
TailwindCSS, and Radix UI, we built an application that is not only technically robust but also
intuitive and user-friendly.

Key accomplishments of this project include:

• Seamless user authentication using Supabase.

• A well-designed dashboard that tracks job statuses.

• A powerful AI-backed slide generation engine that transforms user inputs into structured,
styled slides.

• Real-time feedback through toast noti cations and status updates.

• Support for PDF and PPTX downloads, giving users exibility in how they utilize the
generated content.

• A responsive and themed UI that enhances the overall user experience.

The system has been tested across multiple use-cases like business pitches, educational content, and
short topic summaries, and has shown strong results in all scenarios

9.2 Future Scope

While the current version of Encorp-AI provides a solid foundation, there are several areas for
potential enhancement and expansion:

1. Rich Media Integration

• Support for inserting images, graphs, or charts automatically using AI and APIs (e.g.,
Unsplash or DALL·E).

• Integration with YouTube or stock videos for multimedia-rich slides.

2. Voice Support

• Use of Text-to-Speech (TTS) to allow users to hear speaker notes.

• Option to add voice narration to slides for auto-play presentations.

3. Team Collaboration

Page 42 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

fi
fl
fi
 Page 43 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

31

• Allow multiple users to collaborate in real-time on a single presentation.

• Integration with Google Drive or Notion for shared editing and version control.

4. Presentation Editor

• Introduce a WYSIWYG editor that allows users to tweak fonts, colors, and layout directly
before download.

• Offer drag-and-drop customization of individual slides.

5. Template Marketplace

• Add a feature for selecting or purchasing premium templates, or even allowing users to
upload and sell their own.

6. Multi-language Support

• Enable slide generation in various languages using translation models.

• This would make the tool globally accessible, especially for educational purposes.

7. Mobile App

• A lightweight mobile version (built with React Native or Flutter) could allow users to
generate and review slides on the go.

8. Analytics & Feedback

• Track how users are using their slides: number of downloads, shares, or presentation
feedback.

• Collect anonymized usage data to improve AI performance and UI/UX.

Encorp-AI is not just a presentation generator—it's a step toward democratizing professional

content creation. With the increasing role of AI in education, business, and personal productivity,
this project lays a solid foundation for future innovation. As the platform evolves, it has the
potential to become a full-fledged AI-powered assistant for all types of communication and
storytelling needs.

Page 43 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 44 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

32
REFERENCES

32 The following resources and materials were consulted during the development of the project to
ensure the use of best practices and modern technologies:

8 [1] Next.js, “Next.js Documentation,” Vercel. [Online]. Available: https://nextjs.org/docs.

[Accessed: May 25, 2025].

[2] Tailwind Labs, “Tailwind CSS Documentation,” [Online]. Available: https://tailwindcss.com/

docs. [Accessed: May 25, 2025].

13 [3] OpenAI, “OpenAI Platform Documentation,” [Online]. Available: https://platform.openai.com/

docs. [Accessed: May 25, 2025].

[4] Supabase, “Supabase Documentation,” [Online]. Available: https://supabase.com/docs.

[Accessed: May 25, 2025].

23 [5] Framer, “Framer Motion Documentation,” [Online]. Available: https://www.framer.com/

motion/. [Accessed: May 25, 2025].

5 [6] React Hook Form Team, “React Hook Form Documentation,” [Online]. Available: https://react-
hook-form.com. [Accessed: May 25, 2025].

5 [7] Radix UI, “Radix Primitives Documentation,” [Online]. Available: https://www.radix-ui.com/

docs. [Accessed: May 25, 2025].

5 [8] Recharts, “Recharts Documentation,” [Online]. Available: https://recharts.org/en-US/.

[Accessed: May 25, 2025].

14 [9] B. Brent, “PptxGenJS Documentation,” GitHub Pages. [Online]. Available: https://

gitbrent.github.io/PptxGenJS/. [Accessed: May 25, 2025].

11 [10] jsPDF Contributors, “jsPDF: PDF generation for the browser,” GitHub. [Online]. Available:
https://github.com/parallax/jsPDF. [Accessed: May 25, 2025].

14 [11] Axios, “Axios Documentation,” [Online]. Available: https://axios-http.com/docs/intro.

[Accessed: May 25, 2025].

5 [12] Lucide, “Lucide React Icons,” [Online]. Available: https://lucide.dev/docs/lucide-react.

[Accessed: May 25, 2025].

22 [13] date-fns, “Date-fns Documentation,” [Online]. Available: https://date-fns.org/docs/Getting-

Started. [Accessed: May 25, 2025].

11 [14] GitHub, “Various open-source libraries used in development,” [Online]. Available: https://
github.com. [Accessed: May 25, 2025].

Page 44 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

 Page 45 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

33

APPENDIX

Appendix A – Diagrams

Figure A1: work ow architecture

Appendix B – User Interface Screenshots

Figure B1: Landing page of Encorp-AI

Figure B2: Dashboard

Figure B3: creating form

Figure B4: progress animation

Figure B5: generated pages

Figure B6: Exporting slides/presentation

Figure B7: Exported pdf

Appendix C – Libraries and Packages Used

Package Purpose
next React framework for SSR
tailwindcss CSS utility- rst framework
supabase-js Backend as a Service (BaaS) and Auth
next-auth User authentication
axios HTTP requests
pptxgenjs Generate downloadable PowerPoint les
17 react-hook-form Form management
zod Schema validation
framer-motion Animations

@radix-ui/react-* UI component library

date-fns Date utilities
html2canvas Screenshot DOM for export

Page 45 of 45 - Integrity Submission Submission ID trn:oid:::27535:97857111

fi
fl
fi