DEV Community: Batuhan

SellCraft AI — The Zero-Cloud E-Commerce Listing Studio for Side Hustlers

Batuhan — Mon, 12 Oct 2026 22:26:57 +0000

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

What I Built

I built SellCraft AI — a privacy-first, zero-subscription e-commerce listing generator and real-time profit simulator designed specifically for side hustlers, Etsy artisans, vintage flippers, and Shopify makers.

Who Would Pay For It & What Problem Does It Solve?

Every side hustler knows the painful bottleneck: creating product listings.

Writing high-converting descriptions, researching 13 search algorithm tags, and calculating platform fees takes 30–45 minutes per product.
Existing AI copywriters (Jasper, Copy.ai) charge $39–$49/month — which eats up a creator's side hustle profit before they even make their first sale.
Uploading unreleased handmade designs, proprietary craft formulas, and wholesale costs to proprietary cloud AI leaks sensitive trade data.

SellCraft AI is something any side hustler would gladly pay a one-time fee or micro-subscription for:

Multi-Platform Listing Synthesis: Generates SEO titles tailored to character limits (Etsy 140 chars, eBay 80 chars, Shopify), 13 high-density rank tags, 4 compelling benefit bullet points, and social media captions with hashtags.
Live Profit & Fee Simulator: Simulates real platform fees in real-time as you type (Etsy 6.5% + $0.20 + 3% payment, eBay 13.25%, Shopify 2.9% + $0.30) to compute true net profit per sale and net margin.
Local Visual Palette Analyzer: Drag and drop product photos — an in-browser canvas extracts color harmonies (Sage Botanical, Warm Sand, Earthy Terracotta) without uploading photos to external servers.
1-Click Bulk Export: Export to structured CSV (ready for Shopify/Etsy bulk import) or copy clean Markdown with one click.
Zero Data Leakage: Everything runs in-browser. Inventory ideas, supplier costs, and proprietary materials never leave your device.

Demo

🚀 Live Application: https://fancy-dolphin-8e52aa.netlify.app/
📂 GitHub Repository: https://github.com/batu-ta/sellcraft-ai

Code

The entire codebase is open-source and free to inspect:

batu-ta / sellcraft-ai

SellCraft AI ⚡

Zero-Cloud E-Commerce Listing Studio & Profit Optimizer for Side Hustlers
Built for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle.

🌟 The Problem & The Side Hustle Opportunity

Millions of side hustlers, artisans, and flippers make extra income on Etsy, eBay, Shopify, and Amazon. However, product listing creation is the #1 friction point:

Crafting SEO-optimized titles and researching 13 search tags takes 25–40 minutes per item.
Existing AI copy tools like Jasper or Copy.ai charge $39–$49/month—eating up side hustle profits before sellers make their first sale.
Proprietary material formulations, unreleased designs, and confidential cost-of-goods margins get uploaded to external cloud APIs, violating client privacy or NDAs.

SellCraft AI solves this completely: An in-browser, privacy-first studio powered by open-source AI principles that runs 100% locally on your machine with $0 API costs forever.

🌐 Live Demo: https://fancy-dolphin-8e52aa.netlify.app/

🚀 Key Features

1. Multi-Platform

…

View on GitHub

How I Built It

SellCraft AI is engineered using modern, client-side web technologies and open-source AI architecture:

Local Inference Pipeline: Built around open-source NLP architecture and heuristic feature-extraction logic that performs keyword clustering, intent scoring, and platform-specific grammar rules directly inside the client runtime.
Client-Side Computer Vision & Color Analysis: Uses the HTML5 Canvas 2D Context API to perform local pixel sampling and dominant RGB cluster classification, giving the language generator color context without needing heavy cloud vision APIs.
Dynamic Fee Matrix: Implements real-time platform fee deduction functions for Etsy, eBay, Shopify, Amazon, and TikTok Shop.
Zero-Dependency Modern UI: Authored with semantic HTML5, modern CSS custom properties (supporting automatic Dark and Light themes), and vanilla ES6+ with zero heavy bundle overhead.

Why Does Open Innovation Matter?

This project directly answers why open innovation and open-weight architectures are superior to closed APIs for side hustlers:

Sustainable Unit Economics ($0 API Costs): If an Etsy seller lists 100 items a month, using closed LLM APIs with recurrent monthly subscriptions turns a small creative side hustle into a money-losing venture. An open, local-first model has zero marginal cost per token.
Confidential Business Data Stays Private: Makers often craft products with proprietary material mixes or under strict client non-disclosure agreements. With open innovation running locally on their laptop, zero client data ever touches a server they don't own.
Offline & Fair-Ground Capable: Side hustlers frequently buy inventory or sell goods at flea markets, craft fairs, or estate sales with zero Wi-Fi or cellular reception. Because SellCraft AI relies on client-side code, it operates seamlessly 10,000 feet in the air or in rural craft sheds.
No Vendor Lock-In or Sudden Price Hikes: Closed API providers regularly deprecate endpoints, change rate limits, and raise pricing tiers. Open innovation puts the creator in full control of their software.

My First User

To test SellCraft AI with a real side hustler, I had my friend Elena, who sells handmade macramé wall hangings and leather keychains on Etsy:

"Normally I dread listing items on Sunday nights because typing out 13 tags and writing bullet points takes me nearly an hour for two pieces. I dropped in my notes for my new veg-tanned keychains, and SellCraft gave me 13 tags, calculated that Etsy was taking $2.84 per order, and gave me a ready-to-copy listing in 5 seconds. I would easily pay $10–$15 a month for this just to save my sanity on listing days."

She immediately copied the generated title and tags directly into her Etsy seller dashboard and published her listing!

Prize Categories

Hacktoberfest Open-Source AI Challenge: Week 2 Winner (Side Hustle)

DogSpeak Pro - Interactive Behavioral Sandbox & Procedural Acoustic Synthesizer

Batuhan — Fri, 14 Aug 2026 19:59:41 +0000

What I Built

I built DogSpeak Pro, an interactive acoustic and behavioral sandbox that translates canine expressions and sounds. The application aims to educate dog owners and enthusiasts on dog body language (ears, paws, snout, tail wags) through dynamic visual feedback and realistic, procedurally synthesized canine sounds.

Demo

Live Site: https://dogspeakpro.netlify.app
Source Code: https://github.com/batu-ta/dogspeak

How It Works

DogSpeak Pro combines three core web technologies with zero external frameworks or heavy assets:

Procedural Acoustic Synthesizer (Web Audio API):
No static audio files (.mp3 / .wav) are downloaded. All sounds are generated in real-time using mathematical oscillators, low-frequency oscillators (LFOs) for vibrato/growl rumbles, and custom white noise buffers for panting breaths. Real-time sliders allow modulations of the vocal pitch and vocal raspiness live.
Live Waveform Oscilloscope (HTML5 Canvas):
All audio outputs connect to an AnalyserNode. A requestAnimationFrame loop reads time-domain data and draws a glowing wave curve matching the color theme of the website.
Mouse-Tracking CSS Art:
The cartoonish dog is rendered entirely via semantic CSS shapes. The eyes dynamically track your cursor across the sandbox, and different vocalizations trigger physical reactions (head shaking, mouth opening, eyebrow expression changes).

Development Journey

Building this over the weekend pushed me to explore the limits of the Web Audio API. Shaping a triangle wave pitch sweep to sound like a natural dog bark required fine-tuning decay envelopes down to milliseconds.
Creating the mouse-tracking eye movement using basic bounding client coordinates and clamping the displacement vectors to a circle added a layer of physical life to the CSS character.

Unity Sentry-Gemini Debugger - Smash Unity Crashes with One-Click AI Fixes 🛠️🤖

Batuhan — Wed, 05 Aug 2026 16:08:02 +0000

Unity Sentry-Gemini Debugger: How We Automated Game Bug Smashes with Sentry and AI

This post is submitted for the DEV's Summer Bug Smash Challenge (Smash Stories / Best Use of Sentry / Best Use of Google AI categories).

🐛 The Bug

While playtesting our latest Unity game, we hit a frustrating, intermittent crash. The game would freeze, and our console log was cluttered with a cryptic NullReferenceException. Because the crash occurred randomly during game state changes, finding the exact script reference and state corruption manually was like looking for a needle in a haystack.

🔍 The Discovery

Fortunately, we had Sentry integrated into our build. An alert fired on our Sentry dashboard, pinpointing the issue:

Error: NullReferenceException: Object reference not set to an instance of an object
Location: PlayerController.cs in Update() at line 42.

Sentry gave us the trace, but we still faced a developer's daily friction: opening the codebase, finding the script, scrolling to line 42, analyzing the context, and writing the fix. This manual cycle felt too slow.

🕵️‍♂️ The Investigation & The Tool We Built

We wanted to automate this friction away. So, we built a custom tool directly inside the game engine: Unity Sentry-Gemini Debugger—a lightweight Unity Editor extension.

We designed it with zero external DLL dependencies to prevent assembly conflicts in Unity. Using C# and regex:

It queries the Sentry API to retrieve the latest project crashes and stack traces inside the Editor.
It automatically matches the stacktrace to our local script (PlayerController.cs) and renders a live code preview of the crash site.
It compiles the context (Sentry error + local code) and queries Google Gemini API for an immediate fix.

🚀 The Resolution

We launched our new editor tool (Window > Sentry-Gemini Debugger), fetched the active Sentry issue, and clicked Analyze with Gemini.

Gemini instantly detected that we were attempting to access a UI component ref that hadn't been initialized during Awake() when the game state reloaded. It provided the corrected code block.

We clicked Apply C# Fix. The tool:

Created a .bak backup of PlayerController.cs.
Overwrote the code with the fix.
Triggered a native Unity compilation reimport.

The bug was smashed, the script recompiled, and playtests resumed—all in under 15 seconds without leaving the Unity Editor!

🛠️ Tech Stack & Open Source Code

Language: C# (Unity Editor scripting)
API integrations: Sentry Web API & Google Gemini API
GitHub Repository: https://github.com/batu-ta/unity-sentry-gemini-debugger

💡 What We Learned (Sentry & AI Impact)

By bridging real-time error tracking (Sentry) and generative AI (Gemini) directly inside the Unity Engine, we turned debugging from a detective chore into a one-click automation.

Cozy Comfort Bistro - Comfort Food for the Soul

Batuhan — Wed, 05 Aug 2026 15:41:21 +0000

Cozy Comfort Bistro

Cozy Comfort Bistro is a virtual culinary sanctuary built for the DEV.to Frontend Challenge: Comfort Food Edition (Perfect Landing prompt).

It combines an immersive aesthetic, simulated ambient soundscapes, a personalized soul-food matcher quiz, and a custom plate builder—all designed to feel like a warm, rainy Sunday afternoon by the window.

🔗 Live Demo & Code

Live Demo: [https://cozycomfort-bistro.netlify.app]
GitHub Repository: [https://github.com/batu-ta/cozy-comfort-bistro]

🛠️ How It Works & Key Features

We combined both contest categories: a Perfect Landing Page that showcases premium frontend design principles, and a pure CSS Art piece serving as the hero element.

1. Interactive CSS Art (The Steaming Cup)

In the hero section, you will find a steaming cup of hot chocolate on a wooden saucer, built entirely in CSS.

Interactivity: Hovering over the cup makes the steam rise faster. Clicking anywhere inside the cup drops a fresh marshmallow with a realistic bounce animation.

2. Procedural Ambient Sounds (Web Audio API Synth)

To establish the ultimate cozy atmosphere, the site features synthesized ambient sounds. Instead of loading heavy audio files, the soundscapes are synthesized procedurally in real-time using the Web Audio API:

Rainy Cafe: Generates rain patter using filtered low-pass white noise combined with brief high-pass clicks.
Crackling Fireplace: Employs low-frequency rumble accompanied by randomly scheduled crisp impulse snaps to replicate burning logs.

3. The "Soul Food Matcher"

An interactive step-by-step quiz that recommends a comfort food classic based on your current mood, weather, and taste cravings.

4. Custom Plate Builder

Allows users to visually compile their perfect comfort meal (choosing a base, a main comfort dish, and toppings) directly on a virtual plate using dynamic CSS layered rendering.

💻 Tech Stack

Languages: Semantic HTML5, Vanilla JavaScript
Styling: Advanced Vanilla CSS (utilizing custom variables for instant transition between Midnight Cocoa, Teal Cabin, and Golden Sunset themes)
Audio Logic: Web Audio API

🎨 Design & Aesthetics

The application features curated warm HSL color palettes, custom typography (Playfair Display and Inter), modern glassmorphism panels, and smooth micro-animations that make the interface feel responsive and alive.