DEV Community: Abhimanyu Bhatt

Digital Ofrenda: An AI-Powered Remembrance Altar for Día de los Muertos

Abhimanyu Bhatt — Tue, 27 Oct 2026 13:56:40 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine

What I Built

I built "Digital Ofrenda," an interactive web application designed for Día de los Muertos[cite: 9]. It serves as a digital memorial space where users can honor passed loved ones by lighting virtual candles, placing custom food and memory offerings on an interactive altar, and generating heartfelt poetic remembrance tributes[cite: 9].

Demo

https://snazzy-empanada-94ea31.netlify.app/

https://github.com/abhimanyubhtt-hue/digital-offrenda/tree/main

How I Built It

The app is built around an open-weight model approach, integrating with local Ollama instances running Gemma (http://localhost:11434/api/generate) to stream personalized memorial narratives, complete with a built-in offline fallback generator so the experience never breaks.

Why Does Open Innovation Matter?

Memorializing loved ones deals with deeply personal memories, family stories, and intimate tributes. Utilizing a local open-weight model (Gemma) ensures complete on-device data privacy—keeping sensitive family data strictly local on the user's machine without relying on external, closed cloud APIs.

The Spirit of It

Watching the virtual altar light up with flickering candles, pan de muerto, and marigolds while the local AI weaves a gentle remembrance story brings a comforting, nostalgic warmth that bridges past and present.

Prize Categories

Best Use of Gemma[cite: 4]

SlothMind AI

Abhimanyu Bhatt — Wed, 21 Oct 2026 12:45:45 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady

What I Built

I built SlothMind AI, a privacy-first productivity hub designed for the "Slow and Steady" challenge. It automates tedious, repetitive daily tasks using open-source AI models running completely inside the browser via WebAssembly (@xenova/transformers).

Key features include:

💬 Chat & Email Tidier: Summarizes long, chaotic group chats or email threads into a 2-line executive overview and action checklist.
🗂️ Lazy Note & File Organizer: Instantly cleans up and categorizes messy notes into structured Markdown.
🧊 Fridge-to-Recipe Generator: Turns leftover ingredients into fast, effortless recipe ideas.
🛏️ Wind-Down Companion: Generates calming bedtime stories and ambient soundscapes offline.

Because it runs 100% locally in the browser need nothing.

Demo

https://myapp31.lovable.app

How I Built It

Tech Stack: HTML5, Tailwind CSS, JavaScript (ES6+), Lucide Icons
AI Framework: @xenova/transformers (Transformers.js) running in the browser via WebAssembly (ONNX Runtime)
Deployment: GitHub Pages / Vercel

Development Process

Local-First Architecture: The entire app runs directly in the client's browser. It uses WebAssembly to execute quantized open-source AI models without sending data to any external server.
UI & UX: Designed with a dark glassmorphism theme using Tailwind CSS. Built for maximum sloth-like convenience, featuring pre-loaded sample data so users can test every feature in a single click.
Execution Pipeline:
- Text processing tasks (chat summarizing, file tidying, recipe matching) are executed instantly using light, browser-compatible open-weight models.
- Outputs are structured automatically into actionable lists and Markdown cards with zero manual sorting required. ## Why Does Open Innovation Matter? Running open-source AI locally in the browser provides total data privacy because personal messages and notes never leave the user's device. It eliminates expensive API fees, runs completely free forever, and functions seamlessly without an internet connection while avoiding vendor lock-in.

AI Product Listing Writer for Small Business Sellers 🛍️

Abhimanyu Bhatt — Tue, 13 Oct 2026 13:45:48 +0000

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

What I Built

I built the AI Product Listing Writer, a simple web application that helps small e-commerce sellers (Etsy, eBay, local artisans) generate polished, SEO-optimized product listings instantly!

Demo

https://orange-goggles-96w9jp9rq4j6fxq76-8501.app.github.dev/

Code

https://orange-goggles-96w9jp9rq4j6fxq76.github.dev/

How I Built It

I built the AI Product Listing Writer, a lightweight web application designed to help small e-commerce sellers (Etsy, eBay, local artisans) generate polished, high-converting product descriptions instantly.

The application is built using Streamlit and connects via open inference APIs to open-weight models (such as Qwen 2.5 Coder 32B Instruct). The model takes simple user inputs (Product Name and Key Features) and structures them into SEO-optimized title and description fields for the target platform.

Why Does Open Innovation Matter?

Open innovation and open-weight models give small business owners complete transparency and freedom from locked-in software subscription fees or expensive closed-source API costs. By utilizing open-weight AI models, sellers can process product information locally or via open-weight inference without risking data privacy or facing high operational overhead.

My First User

Touch Grass AI

Abhimanyu Bhatt — Thu, 08 Oct 2026 13:51:48 +0000

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

What I Built

i created streamlit application to allow individuals to upload picture of plants and birds or trails. it encourage people to explore outdoors by uploading image.

Demo

this app runs locally with streamlit to help people explore outdoors identify birds and plants directly from photos.

Code

https://github.com/abhimanyubhtt-hue/touch-grass-ai/tree/main

How I Built It

using python and streamlit to give people access to details just by uploading image directly and it runs locally.

Why Does Open Innovation Matter?

open source ai ensure nature identifying tools with a constant internet(offline) allowing people to explore outdoors.

Prize Categories

TOUCH GRASS CHALLENGE

hi

Abhimanyu Bhatt — Wed, 07 Oct 2026 13:52:26 +0000

hi

Abhimanyu Bhatt — Wed, 07 Oct 2026 13:43:19 +0000