Vidya – AI Study Platform

A cinematic, AA-compliant learning sanctuary—where 'inclusive education' isn't just a checklist, but a multimodal AI exp...

LifecycleAug 2025 - Present
ProtocolWebsite
Vidya – AI Study Platform
Narrative Exhibit
Project Genesis

Why and How
I built this.

"I realized that the 'Big Gap' in education is the rigidity of the medium. Textbooks are often 'locked rooms' for students with disabilities."

"Collaborated with accessibility experts to map the 'Inclusive Learning Pathway,' auditing LMS platforms for AA compliance."

"We validated that students with disabilities spend up to 3x more time simply accessing content than learning it."

01Problem Identification

Traditional education has a gatekeeping problem. While most students can simply open a PDF or watch a YouTube lecture, over 10 million students with visual or hearing impairments, or learning disabilities like dyslexia, are met with an invisible wall.

Educational materials are fundamentally designed for the average learner, leaving students with disabilities to struggle with inaccessible layouts, complex jargon, and non-adaptive formats.

Existing 'accessibility tools' are often clunky and lack the premium aesthetic needed to make learning truly effortless.

02The Solution Hypothesis

Vidya acts as a universal translator for knowledge. It 'inhales' any medium—a complex PDF, a fast-paced video, or messy handwritten notes—and 'exhale' it in the specific format a student needs: be it audio, simplified text, or interactive quizzes.

03The "Oops" Moment

In the early pilot, I focused too much on making the AI 'smart' and not enough on making the UI 'simple.' The dashboard was overwhelmed by complex menus.

Testing with a screen-reader user revealed 'navigational noise.' I realized that true accessibility is minimalism.

I stripped the UI back to essentials, implementing the 'Single Action' rule where the primary learning node is always the main focus.

Inclusive Learning Pathway mapping
Multimodal Transformation Engine
Screen-reader optimized UI architecture
High-contrast 'Electric Pulse' visual theme
Real-time AI-driven pedagogical summaries
Strategic Analysis

Modern Discovery.

Conducted an intensive discovery phase involving focus-group participants and competitive landscape audits to establish the project's foundation.

Core Findings

Strategic Breakthroughs.

Finding 01

Users feel overwhelmed by data complexity

Finding 02

Mobile-first interaction is non-negotiable

Finding 03

Speed-to-value determines long-term retention

Research Methods

Strategic Methodologies.

Phase 01
Core Method
User Surveys
Phase 01
Core Method
Hotjar Analysis
Phase 01
Core Method
A/B Testing
Phase 01
Core Method
Focus Groups
System Architecture

Multimodal Transformation Architecture.

Inclusive learning pathway with accessibility guards

01
Multimodal Source

PDF documents, audio, video & web text sources

02
Accessibility Adaptor

Decision: Checks font scale, contrast & screen reader rules

03
Gemini AI Transformer

Generates bite-sized summaries & adaptive quizzes

04
Inclusive Learning Hub

Accessible interactive study workspace

”
Core Logic & Architecture

"Vidya ingests complex educational media (PDFs, videos, handwritten notes) and filters them through AA/AAA accessibility rules. The system exhales content into Dyslexia-friendly typography, voice guidance, and bite-sized interactive quizzes."

Workflow Evolution

Design
Process.

The core of Vidya is the Inclusive Learning Pathway. In my sketchbook, I mapped out the 'Multimodal Engine,' tracing how a source is processed into specific formats like Audio, Simplified Text, or Quizzes based on learned needs.

State 01

Inhale Phase: A PDF or YouTube Link is ingested by the Multimodal Engine.

State 02

Logic Branching: The AI determines the necessary transformation (Audio, Dyslexia-friendly text, or Summary).

State 03

Exhale Phase: The content is delivered as a screen-reader optimized summary or voice-guided interactive quiz.

State 04

Validation: Pedagogy is checked to ensure simplified content retains core academic meaning.

Workflow Mapping Sketch
Visual Protocol

Color
Palette.

The design philosophy is 'Inclusive Clarity.' We balanced strict accessibility standards with a modern, high-end feel.

Electric AI Pulse (#0284C7) represents focus and multimodal energy, while Deep Obsidian (#0D1117) ensures AAA color contrast ratios alongside calming teal accents (#2DD4BF).

Typography choice: Outfit for premium headers and OpenSans for dyslexic-friendly body text readability.

Typography & Fonts

Outfit (Headers)

OpenSans (Body/Dyslexia-friendly)

Signature Atoms
Pedagogical NodeMultimodal EngineElectric Pulse Mode
#0284C7
#0D1117
#2DD4BF
Logic & Infrastructure

Engineering
Excellence.

Built with Next.js for SSR and performance. The Multimodal AI Engine leverages the Gemini API for pedagogical summarization.

Integrated OpenAI Whisper for video transcription and OpenAI Vision for complex document analysis.

Developed a real-time engine that generates AI-driven quizzes based on transformed content to boost retention.

Frontend Layer

Built with Next.js & Tailwind for lightning-fast rendering and responsive boutiques.

Core Engine

Leverages PostgreSQL & AI SDKs to process complex user intents and data structures.

Next.js
Gemini AI
OpenAI Vision
Tailwind CSS
React
Production Ready

Measured Impact.

95%

95% accuracy in pedagogical content simplification

35%

Increased student engagement by 35% via interactive quizzes

10M+

Designed a pathway capable of serving 10M+ diverse learners

Back to All Works
FIRST CLASS
BOARDING PASS
CLASS: FIRST CLASS
Name of Passenger

Sakshi Agrahari

FromAI PRODUCTDESIGN ENGINEER
ToYOUR TEAMCOLLABORATION
FlightSA-2026
DateANYTIME
GateW.W.W
Seat01A
CONNECT WITH ME // SOCIALS
BOARDING PASS
TKT-82930-SA
© 2026 BUILT FOR IMPACT
Cinnamon Roll Character