DEV Community: Snehal Ghadge

freelance ai

Snehal Ghadge — Wed, 21 Oct 2026 14:13:45 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle

What I Built---

title: "FreelanceFlow AI: 100% Private, Zero-Cost Invoice & Expense Copilot Powered by Local Gemma & Ollama"
published: true
tags: devchallenge, hf26challenge, python, opensource

canonical_url: false

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle

What I Built

FreelanceFlow AI is a 100% local, privacy-first invoice and expense manager designed specifically for freelancers, solopreneurs, and independent contractors.

The Problem

At the end of every month or tax season, freelancers face a painful chore: hunting down dozens of digital receipts and PDF invoices scattered across email inboxes and downloads folders (GitHub Copilot, Figma, AWS, Vercel, WeWork, Zoom, client dinners, and hardware purchases). Today, they have two frustrating choices:

The Manual Spreadsheet Grind: Manually typing invoice numbers, dates, tax amounts, and vendors into Excel or Google Sheets. It takes hours, leads to typos, and frequently causes missed tax write-offs.
Expensive Cloud SaaS Apps: Paying $20–$50/month for tools like Expensify, Dext, or QuickBooks that upload confidential financial documents, client names, and bank statements to proprietary cloud LLM APIs.

Who Would Use & Pay for It?

Freelancers and solopreneurs running lean side hustles who care deeply about:

Financial Data Confidentiality: Keeping client billing records, NDAs, and receipts strictly on their own machines.
Eliminating Recurring SaaS Fees: Paying zero monthly AI token or subscription fees.
Tax Readiness: Getting clean, categorized summaries and 1-click CSV exports ready for tax deductions and accountant handoffs.

Demo

The app features a modern, responsive SaaS-style interface powered by Streamlit and Plotly:

📊 Financial Health Dashboard: 4 real-time KPI metric cards (Total Spend, Invoices Logged, Average Expense, and Top Category), plus interactive Plotly visualizations for monthly spending trends and category distribution.
📄 Smart Document Ingestion: Drag-and-drop support for PDF invoices and receipt images (PNG, JPG, WEBP).
🛡️ Instant Duplicate Detection: Computes a cryptographic SHA-256 hash on upload and alerts you immediately if a document was already entered, preventing double deductions.
🤖 Gemma AI Extraction with Human-in-the-Loop: Automatically parses Vendor, Invoice #, Date, Currency, Subtotal, Tax, Total, and Category into an editable confirmation form.
💼 Full Expense Manager: Filter by category, search by vendor/notes, adjust records in-place, and export clean CSVs.
🧪 1-Click Demo Mode: Includes 12 pre-loaded fictional freelance expenses (Figma, GitHub, AWS, WeWork, Zoom, etc.) so judges can test every feature instantly without needing an active Ollama instance.

A ready-to-test sample invoice (sample_invoice.pdf) is included in the repository.

Code

snehalghadge59-blip / freelanceflow-ai

💼 FreelanceFlow AI — Private AI Expense & Invoice Copilot

Built for the Hacktoberfest 2026 Week 2 Side Hustle challenge (#devchallenge, #hf26challenge).

FreelanceFlow AI is a 100% local, privacy-first invoice and expense manager designed specifically for freelancers, solopreneurs, and contractors. It turns messy PDF invoices and receipt images into structured, categorized tax-ready expense records using a locally running Google Gemma model served through Ollama and a local SQLite database.

🌟 Why FreelanceFlow AI?

As a freelancer, keeping track of deductible business expenses usually means juggling PDF receipts, taking screenshots of SaaS invoices, and spending hours each quarter manually typing numbers into spreadsheets.

Cloud AI alternatives exist, but they have major drawbacks:

Financial Privacy Risks: Uploading bank statements, NDA-bound invoices, and personal receipts to third-party cloud LLM APIs poses severe confidentiality issues.
Recurring Token Costs: Monthly SaaS fees and API bills eat away at freelance margins.
Vendor Lock-In: Your expense history…

View on GitHub

The repository includes the full source code, test suite, and Windows setup instructions:

app.py: Modern SaaS UI, interactive charts, and review forms
database.py: Parameterized SQLite queries, duplicate hash checks, and demo data generator
ai_service.py: Local Ollama REST client, Gemma prompt engineer, and Pydantic validation
invoice_parser.py: PyMuPDF native PDF text extraction, Tesseract OCR wrapper, and SHA-256 hashing
tests/: 13 passing unit tests verifying ACID compliance, Pydantic normalization, and document parsing

How I Built It

FreelanceFlow AI is built entirely on open-source technologies and local-first architecture:

[ PDF / Image Receipts ]
           │
           ▼
  [ invoice_parser.py ] ──► Computes SHA-256 Hash ──► Duplicate Check
  ├── PyMuPDF (Native PDF Text Extraction)
  └── Tesseract OCR (Image Text Extraction)
           │
           ▼ Raw Document Text
    [ ai_service.py ]
  └── Ollama REST API (http://localhost:11434)
        └── Google Gemma 2 (gemma2:2b / gemma:2b)
           │
           ▼ Strict JSON Extraction
    [ Pydantic Validation ]
  └── Cleans currency symbols ($/€/£), formats dates (YYYY-MM-DD), validates totals
           │
           ▼
    [ Streamlit UI (app.py) ] ◄── Human Verification & Edit
           │
           ▼ Parameterized SQL (Zero Injection)
    [ SQLite database.py ] ──► Plotly Visualizations & CSV Export

Key Components:

Open-Weight AI Model: Google's Gemma 2 (gemma2:2b) served locally via Ollama. Gemma's compact size and high reasoning quality allow it to reliably extract structured financial JSON on standard laptops without requiring a discrete GPU.
Deterministic Data Validation: All LLM outputs are piped into a strict Pydantic model (InvoiceExtractionResult) that normalizes currency symbols, enforces YYYY-MM-DD ISO dates, rounds monetary figures, and maps vendors to standardized tax expense categories.
Local Text Extraction: PyMuPDF (pymupdf) extracts clean text from multi-page PDF invoices in milliseconds, while Pillow + Tesseract OCR provides scanned receipt parsing.
Relational Storage: SQLite with parameterized queries (?) guarantees data safety and SQL injection protection.

Why Does Open Innovation Matter?

Open innovation is essential for personal financial software:

Complete Data Sovereignty & Confidentiality: Financial receipts contain sensitive client names, business addresses, tax IDs, and bank account snippets. Sending this data to closed cloud APIs risks leakage and unauthorized training on user financial habits. Open-weight models like Gemma run entirely on your local machine with zero external network requests.
Zero Marginal Operating Costs: Closed AI APIs charge per token and impose rate limits. Open-source local models allow freelancers to parse thousands of invoices completely free of charge.
Offline Reliability: Freelancers frequently travel and work from airplanes, trains, or remote spots with intermittent connectivity. Because both SQLite and Gemma run locally, FreelanceFlow AI works seamlessly completely off-grid.
Resilience to Deprecation: Closed APIs change pricing, modify model behaviors, or shut down endpoints at will. With open-source code and local weights, your expense tool will function predictably for years.

My First User

I shared the MVP with a freelance web designer who typically accumulates 15–20 software and client-meeting receipts every month.

They tested it with their recent invoices (DigitalOcean hosting, Figma subscription, and an ergonomic keyboard purchase). Their feedback:

"The duplicate warning alone saved me from logging the same $12 cloud invoice twice. I love that it doesn't require a monthly subscription like QuickBooks, and having an instant CSV ready for tax prep at the end of the year is a lifesaver."

Prize Categories

Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle

Demo

Code

How I Built It

Why Does Open Innovation Matter?

My Agent Session

My First User

Prize Categories

FreelanceFlow AI

Snehal Ghadge — Sun, 18 Oct 2026 17:40:30 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle

What I Built

FreelanceFlow AI is a 100% local, privacy-first invoice and expense manager designed specifically for freelancers, solopreneurs, and independent contractors.

The Problem

The Manual Spreadsheet Grind: Manually typing invoice numbers, dates, tax amounts, and vendors into Excel or Google Sheets. It takes hours, leads to typos, and frequently causes missed tax write-offs.
Expensive Cloud SaaS Apps: Paying $20–$50/month for tools like Expensify, Dext, or QuickBooks that upload confidential financial documents, client names, and bank statements to proprietary cloud LLM APIs.

Who Would Use & Pay for It?

Freelancers and solopreneurs running lean side hustles who care deeply about:

Financial Data Confidentiality: Keeping client billing records, NDAs, and receipts strictly on their own machines.
Eliminating Recurring SaaS Fees: Paying zero monthly AI token or subscription fees.
Tax Readiness: Getting clean, categorized summaries and 1-click CSV exports ready for tax deductions and accountant handoffs.

Demo

The app features a modern, responsive SaaS-style interface powered by Streamlit and Plotly:

📊 Financial Health Dashboard: 4 real-time KPI metric cards (Total Spend, Invoices Logged, Average Expense, and Top Category), plus interactive Plotly visualizations for monthly spending trends and category distribution.
📄 Smart Document Ingestion: Drag-and-drop support for PDF invoices and receipt images (PNG, JPG, WEBP).
🛡️ Instant Duplicate Detection: Computes a cryptographic SHA-256 hash on upload and alerts you immediately if a document was already entered, preventing double deductions.
🤖 Gemma AI Extraction with Human-in-the-Loop: Automatically parses Vendor, Invoice #, Date, Currency, Subtotal, Tax, Total, and Category into an editable confirmation form.
💼 Full Expense Manager: Filter by category, search by vendor/notes, adjust records in-place, and export clean CSVs.
🧪 1-Click Demo Mode: Includes 12 pre-loaded fictional freelance expenses (Figma, GitHub, AWS, WeWork, Zoom, etc.) so judges can test every feature instantly without needing an active Ollama instance.

A ready-to-test sample invoice (sample_invoice.pdf) is included in the repository.

Code

snehalghadge59-blip / freelanceflow-ai

💼 FreelanceFlow AI — Private AI Expense & Invoice Copilot

Built for the Hacktoberfest 2026 Week 2 Side Hustle challenge (#devchallenge, #hf26challenge).

🌟 Why FreelanceFlow AI?

Cloud AI alternatives exist, but they have major drawbacks:

Financial Privacy Risks: Uploading bank statements, NDA-bound invoices, and personal receipts to third-party cloud LLM APIs poses severe confidentiality issues.
Recurring Token Costs: Monthly SaaS fees and API bills eat away at freelance margins.
Vendor Lock-In: Your expense history…

View on GitHub

The repository includes the full source code, test suite, and Windows setup instructions:

app.py: Modern SaaS UI, interactive charts, and review forms
database.py: Parameterized SQLite queries, duplicate hash checks, and demo data generator
ai_service.py: Local Ollama REST client, Gemma prompt engineer, and Pydantic validation
invoice_parser.py: PyMuPDF native PDF text extraction, Tesseract OCR wrapper, and SHA-256 hashing
tests/: 13 passing unit tests verifying ACID compliance, Pydantic normalization, and document parsing

How I Built It

FreelanceFlow AI is built entirely on open-source technologies and local-first architecture:

🌿 TactileTrail — Zero-Screen Audio Naturalist

Snehal Ghadge — Sun, 11 Oct 2026 12:45:19 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass

🌿 TactileTrail — Zero-Screen Audio Naturalist

Put your phone in your pocket. Put your earbuds in. Step into the living world.

What I Built

Traditional "nature and plant identification apps" suffer from a tragic irony: they force you to stare at a glowing glass rectangle while standing in the middle of a forest. Hikers and strollers wander through pristine woodlands with cameras glued to their screens, reading digital text about leaves instead of touching the bark directly in front of them.

TactileTrail flips this paradigm completely.

TactileTrail is a Zero-Screen Audio Naturalist built for anyone yearning to genuinely reconnect with the outdoors. It was designed from the ground up to get your eyes off the screen and your senses engaged with the living earth:

Pick Your Outdoor Biome: Choose between Forest Trail, Botanical Park, Neighborhood Stroll, River Shore, or Open Meadow.
Select Walk Length & Mode: Choose short (3 checkpoints, ~4 min) or standard (4 checkpoints, ~6 min). Select Cloud AI or 100% Offline Edge Mode.
Slide Your Phone into Your Pocket: Put on your earbuds, lock your screen, and walk.

How It Liberates You from the Screen:

Intimate Voice of Sir David Attenborough: Spoken natural history micro-quests guide your attention to specific botanical phenomena (e.g., rhytidome cork ridges, psithurism canopy friction, stomata venation, geosmin scent).
Physical Haptic Feedback (navigator.vibrate): You never have to look at your phone. Gentle rhythmic vibrations in your pocket tell you when to pause, close your eyes, reach out to touch a leaf, or take a deep grounding breath.
Hardware Earbud Controls (HTML5 MediaSession API): Use your headphone buttons (Play/Pause/Next track) or lock-screen controls to advance checkpoints without taking your device out of your pocket.
Pocket Shield: A dedicated OLED blackout mode prevents accidental thigh taps while walking.
Sunlight High-Contrast Mode: High-visibility emerald and white layout designed specifically for readability under direct outdoor sun glare.
Post-Walk Field Journal: When you finish, review your mindful time outdoors, snap a quick photo note of a specimen you touched, and record field reflections.

Demo

Live Repository: https://github.com/snehalghadge59-blip/TactileTrail
Deployment Blueprint: Configured for instant deployment on Render via render.yaml & Procfile.

Walkthrough & Experience Progression

Checkpoint	Sense	Field Action	Botanical & Physical Phenomenon
1. The Cartography of Bark	🖐️ Touch	Tracing furrowed ridges with closed eyes	Rhytidome cork formation, phloem vascular sap, and north-facing moss cushions
2. Acoustic Canopy Layers	👂 Sound	Dissecting foliage frequencies in 3D	Psithurism (wind-leaf friction) and avian frequency modulation
3. Capillary Architecture	👁️ Observation	Backlighting a fallen leaf against the sun	Fractal vein geometry, xylem negative hydrostatic tension, stomata
4. Loamy Grounding	🧘 Reflection	Deep inhalation of damp soil and air	Geosmin soil actinobacteria & conifer phytoncides

graph TD
    User([🚶 Outdoor Walker]) -->|1. Selects Biome & Pockets Phone| UI[📱 TactileTrail Web App]

    subgraph Open-Source AI Intelligence
        UI -->|Generate Quests| Router{Inference Router}
        Router -->|Cloud Mode| GAI[Google AI Studio - Gemma / Gemini API]
        Router -->|Off-Grid Wilderness Mode| Edge[⚡ Local Gemma Edge Engine]
    end

    subgraph Sensory Audio Synthesis
        UI -->|Narration Stream| AudioRouter{Audio Pipeline}
        AudioRouter -->|Neural Voice| EL[🎙️ ElevenLabs Naturalist TTS]
        AudioRouter -->|Local Server Fallback| GTTS[🔊 gTTS Audio Engine]
        AudioRouter -->|Browser Fallback| WebSpeech[🗣️ Web SpeechSynthesis API]
        UI --> Soundscape[🍃 Procedural Web Audio API Soundscape]
        Soundscape -->|Wind Gusts & Avian Chirps| Earbuds([🎧 Earbuds / Headphones])
    end

    subgraph Zero-Screen Pocket Interaction
        UI --> MediaSession[📻 HTML5 MediaSession API]
        MediaSession -->|Earbud Clicks: Next/Pause/Play| UI
        UI --> Haptics[📳 Mobile Vibration Haptics]
        UI --> PocketMode[🔒 Pocket Shield Blackout]
    end

    UI -->|Walk Complete| Journal[📔 Field Journal & Specimen Photo Note]

Code

snehalghadge59-blip / TactileTrail

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass

🌿 TactileTrail — Zero-Screen Audio Naturalist

Put your phone in your pocket. Put your earbuds in. Step into the living world.

What I Built

TactileTrail flips this paradigm completely.

Pick Your Outdoor Biome: Choose between Forest Trail, Botanical Park, Neighborhood…

View on GitHub

GitHub Repository URL: https://github.com/snehalghadge59-blip/TactileTrail

Repository Structure

TactileTrail/
├── app.py                  # FastAPI server with Google Gemma, ElevenLabs & Gemma Edge engine
├── test_app.py             # Complete automated integration test suite (7/7 passing)
├── requirements.txt        # Production dependencies (FastAPI, uvicorn, httpx, gTTS)
├── .env.example            # Environment variables configuration template
├── Procfile                # Render web process definition
├── render.yaml             # Render infrastructure as code blueprint
├── README.md               # Challenge submission & documentation
├── templates/
│   └── index.html          # Responsive mobile outdoor UI with Pocket Shield
└── static/
    ├── css/
    │   └── style.css       # Outdoor design system (Dark Forest & Sunlight modes)
    └── js/
        └── app.js          # Procedural Web Audio soundscape & MediaSession hardware hooks

How I Built It

1. Open-Source AI Engine: Google Gemma 2

TactileTrail utilizes Google Gemma (via Google AI Studio and Hugging Face Serverless Inference API). Gemma's calibrated system prompt is instructed to roleplay an intimate, scientifically rigorous field naturalist in the spirit of David Attenborough:

Prompts emphasize tactile botanical vocabulary: phloem, rhytidome cork, psithurism, geosmin, capillary tension, basal meristem.
Strictly caps each checkpoint narration to 50–75 spoken words to avoid cognitive overload while walking.
Outputs structured JSON specifying tactile invites, haptic cues (pulse, double, gentle), and estimated duration.

2. Off-Grid Gemma Edge Engine (100% Offline Inference)

Recognizing that cellular service disappears in deep forests and mountain passes, TactileTrail includes a localized Gemma Edge Engine. It emulates Gemma’s prompt outputs with deterministic, parametric natural history repositories for all five biomes, allowing complete trail immersion with zero connectivity.

3. Procedural Web Audio API Nature Soundscape

Streaming heavy audio files fails completely without internet. Instead of MP3 soundscapes, TactileTrail synthesizes nature in real time using the browser's Web Audio API:

Wind Gusts (Psithurism): Filtered pink noise modulated with dual low-frequency oscillators (LFOs) mimicking canopy wind swells.
Avian Birdsong: Algorithmic frequency-modulated (FM) oscillators generating spaced songbird chirps.
Zero data footprint: Generates infinite ambient nature soundscapes with 0 KB bandwidth consumption.

4. Resilient Multi-Tier Audio Narration Pipeline

Spoken checkpoint quests are rendered with a triple-redundant architecture:

Primary: ElevenLabs Neural TTS for realistic, comforting naturalist pacing.
Server-Side Fallback: Local gTTS (Google Text-to-Speech) engine streaming MP3 bytes.
Client-Side Fallback: Web SpeechSynthesis API with tuned pitch (0.94) and cadence (0.88), functional even on an airplane or deep in the backcountry.

5. Zero-Screen Hardware Integration

HTML5 MediaSession API: Registers track metadata (title, artist, album) and hooks hardware earbud click events (nexttrack, previoustrack, play, pause) to step through checkpoints.
Vibration API: Vibrates your pocket when an action is required so you never need to look down.

Why Does Open Innovation Matter?

The "Trail Signal Paradox"

When you hike three miles into an old-growth cedar grove, descend into a sandstone canyon, or climb a ridge, cellular signal drops to zero.

Proprietary, closed-source, cloud-only AI models (such as closed cloud APIs) fundamentally fail in genuine wilderness. If an AI naturalist requires a gigabit connection to an enterprise data center, it is completely useless where nature actually lives.

Why Open-Weights Models (Gemma) Win:

True Edge Sovereignty: Open-weight models like Google Gemma can be quantized (GGUF, llama.cpp, WebGPU) and run directly on a hiker's mobile phone, laptop, or low-power field device.
Zero Latency on the Trail: On-device edge weights eliminate network round-trips, delivering immediate guidance the moment you reach a clearing.
Privacy in Solitude: Walking in nature is an intimate, contemplative practice. Open models running locally ensure that personal field journals, audio notes, and location data never leave the user's device.
Domain Fine-Tuning: Open weights empower naturalists, ecologists, and park services to fine-tune specialized models for regional flora and fauna without vendor lock-in or recurring API costs.

My Agent Session

This project was built and iterated using Google Antigravity IDE:

Designed the end-to-end architecture connecting FastAPI, Google AI Studio, Gemma, and ElevenLabs.
Crafted the procedural Web Audio synthesizer algorithms for wind and bird frequencies.
Integrated the HTML5 MediaSession hardware button handlers and Pocket Shield touch locks.
Validated all 5 biomes and fallback pathways with an automated test suite (python test_app.py).

Prize Categories

Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass (Primary Track)
Google Gemma / Open-Weights AI Track

StudyPulse — Your All-in-One Study Buddy

Snehal Ghadge — Fri, 02 Oct 2026 16:19:16 +0000

StudyPulse — Your All-in-One Study Buddy
This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend

What I Built
I built StudyPulse, a minimalist, distraction-free, and privacy-first study companion designed specifically for my best friend Alex, who is juggling college coursework, coding bootcamps, and studying for technical certifications.

The Problem It Solves
Alex constantly struggled with three major pain points while studying:

Context Switching Fatigue: Hopping between a timer app on their phone, Anki for flashcards, and a quiz website on their browser broke their flow and opened the door to endless social media distractions.
Clunky & Rigid Tools: Most flashcard and quiz apps are overcomplicated, filled with ads, require accounts/subscriptions, and lack a built-in focus rhythm.
Loss of Flow: Whenever Alex switched tabs to review a flashcard, phone timers would beep unexpectedly or get lost in the background.
The Solution
StudyPulse consolidates the entire study cycle into one unified, elegant, and responsive single-page web app:

⏱️ Customizable Pomodoro Clock: A visual circular countdown timer with Focus, Short Break, and Long Break sessions, quick +5m / -5m adjustments, session streak counter, and persistent mini-timer widget so you never lose track of time when navigating tabs.
🎧 Focus Ambient Synthesizer: Pure Web Audio API-synthesized soothing soundscapes (Gentle Rain, Soft White Noise, and 10Hz Alpha Waves binaural beats) with zero external audio assets or network dependencies.
🗂️ Interactive 3D Flashcards: Tactile card flip animations with keyboard controls (Space to flip, 1/2/3 to rate mastery), deck management, and progress tracking.
📝 Automated Quiz Master: Automatically turns any flashcard deck into an active-recall multiple-choice quiz with smart distractors, custom question creation, timers, immediate feedback with concept explanations, and celebratory confetti upon achieving mastery.
🔒 100% Offline & Private: Zero tracking, zero logins, and no servers required. All data stays in the browser's localStorage, with one-click JSON Export & Import to backup study decks.
Demo
Live Local Preview: Run python -m http.server 8080 or open https://snehalghadge59-blip.github.io/studdy-buddy/ directly in any modern web browser.
Recording & Screenshots: Check the project demo showing the Pomodoro clock running seamlessly alongside 3D flashcards and automated quiz scoring.
Key Screenshots & Flow:
Pomodoro Focus Dashboard: Modern glassmorphic dark theme, circular progress ring, and customizable session intervals.
3D Flashcards Deck: Smooth flip physics, keyboard navigation, and spaced-repetition mastery status.
Quiz Master Challenge: Automated multiple-choice generator with timer options, instant correctness feedback, and confetti celebration.
Settings & Backup: Custom interval adjustment, audio volume toggles, and JSON export/import.
Code
The codebase is written with clean, standard web technologies without heavy dependencies:

index.html: Accessible, semantic HTML5 structure with SVG icons and dialog modals.
style.css: Custom design system featuring CSS variables, dark/light themes, 3D CSS transforms (perspective: 1200px), and responsive layouts.
app.js: Modular, dependency-free vanilla JavaScript:
SoundManager: Native Web Audio API synthesizer for acoustic chimes and ambient sound generators.
AppState: LocalStorage state management with streak tracking.
PomodoroController: Timer engine with SVG circle math and tab synchronizer.
FlashcardController: 3D card flip engine, mastery tracking, and deck CRUD.
QuizController: Distractor generator, question runner, and score evaluator.
Confetti: Canvas-based physics particle engine for celebratory animations.
How I Built It
To build StudyPulse rapidly over the weekend, I utilized Agentic AI pair-programming with the DeepMind Antigravity framework running autonomous subagent cycles for end-to-end development:

Iterative Scaffolding: Generated a modular, self-contained architecture focused on performance, accessibility, and zero-runtime overhead.
Web Audio Synthesizer: Rather than relying on external MP3 audio files that might fail to load or get blocked by ad-blockers, the AI engineered a native Web Audio API acoustic chime and pink noise / binaural beat oscillator directly in JavaScript.
Autonomous Browser Subagent Verification:
The agent booted up a headless browser instance at http://localhost:8080.
Verified that the timer starts, updates the circular SVG progress ring, and toggles play/pause states.
Verified 3D flashcard card flipping, keyboard shortcut detection (Space, 1, 2, 3), and mastery percentage updates.
Tested the Quiz Master question generator, distractor assignment, timer countdown, and answer review mechanisms.
Tested settings customization and local storage persistence.
Why Does Open Innovation Matter?
Open innovation and open web standards are the foundation of accessible education:

No Paywalls or Subscriptions: Learning tools should be free, open-source, and accessible to anyone with a browser—whether on a smartphone, tablet, or laptop.
Privacy First: With closed SaaS study platforms, student study habits, personal notes, and data are tracked and monetized. With open-source and local-first architecture, the user has 100% data ownership and privacy.
Offline & Resilient: Open web standards ensure this app functions anywhere—even on a train or off-the-grid without an active internet connection.
Community Hackability: Anyone can fork this project, add new card export formats (Anki, Quizlet, CSV), or tune the ambient audio frequencies to their own study style.
My Agent Session
Tooling Used: Antigravity IDE, Browser Subagent Automation, Web Audio API, Vanilla HTML5/CSS3/ES6.
Workflow Highlights: Autonomous design prototyping, real-time browser interaction validation, and zero-friction build pipeline.
Prize Categories
Primary: Hacktoberfest Weekend Challenge: Build for a Friend
Theme: Productivity, Education, and Accessible Web Tools