DEV Community: Sai Rathod

Hacktoberfest Open-Source AI Challenge: Week 2

Sai Rathod — Thu, 15 Oct 2026 08:41:32 +0000

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

What I Built

SellerPilot AI is an AI-powered inventory management application designed for small retailers and online sellers.

Small businesses often struggle to track stock, identify best-selling products, and decide when to reorder inventory. SellerPilot aims to simplify these tasks through a single dashboard.

Key features include:

Inventory and product management.
Sales tracking and analytics.
Best-selling and slow-moving product reports.
Smart restocking recommendations.
AI-powered product description generation.
Local data storage and backup capabilities.

The goal is to help small business owners make better inventory decisions without depending on a permanent internet connection.

Demo

GitHub Repository: https://github.com/CronixDev09/SellerPilotAi

A video demonstration will be added once the demo recording is ready.

Code

Explore the source code on GitHub.

How I Built It

I built SellerPilot AI using React, TypeScript, Vite, Tailwind CSS, Electron, and SQLite.

The application is designed around local data storage and offline-first inventory operations.

For AI functionality, the project integrates with Ollama to run compatible open-weight models, including Gemma. The AI assists with product descriptions and natural-language explanations of business insights.

Inventory calculations, sales totals, and restocking recommendations use application logic and stored data rather than relying on the language model to generate numbers.

Why Does Open Innovation Matter?

Small businesses need control over their business data and operating costs.

Using open-weight AI makes local inference possible, reducing the need to send business information to external AI services. It also gives the project flexibility to experiment with different compatible models.

Local inference avoids per-request fees from hosted AI APIs, although users still need suitable hardware.

For SellerPilot, open innovation provides an opportunity to combine AI assistance with local data ownership and offline functionality.

My Agent Session

I used AI-assisted development to build and refine the application. An agent session recording will be added if available.

My First User

My next step is to test SellerPilot with a real shop owner, gather feedback, and evaluate whether its inventory and restocking features solve a meaningful business problem.

Prize Categories
Best usecase of openweight models

"Build for a Friend" challenge

Sai Rathod — Wed, 07 Oct 2026 14:53:47 +0000

Summary of What Was Created:

Challenge Template Alignment: Follows the official Dev.to Hacktoberfest "Build for a Friend" challenge structure.

GitHub Repository Link Included: Added both the direct link

CronixDev09 / Code-Coach

CodeCoach is an AI-powered DSA learning coach for beginners, using Gemma to explain concepts, analyze Python code, provide progressive hints, and help students solve problems independently.

🎓 CodeCoach

AI-Powered DSA Learning Coach for Beginners (Python)
Built for Hacktoberfest — "Build for a Friend"
Intelligence powered by the Gemma open-weight model

💡 1. Product Overview

CodeCoach is a personalized AI coding tutor designed for a beginner learning Data Structures & Algorithms using Python.

Most beginners understand DSA definitions (e.g., "an array stores multiple values"), but freeze when they face an empty editor and have to translate that theory into working code. Traditional platforms either dump the full solution or generic chatbots provide copy-paste code, depriving the student of the reasoning and discovery experience.

Core Principle:

"Don't solve for the student. Teach the student how to solve."

🏆 2. The Hackathon Challenge: "Build for a Friend"

CodeCoach is built for a real friend: Rahul, a college sophomore who understands theoretical concepts but struggles with:

0-based array indexing and IndexError: list index out of range
Loop bounds…

View on GitHub

What I Built: The story of building CodeCoach for friend, the theory-to-code problem, core features, and real student testimonial.
Demo: Demo route /demo walkthrough and local startup instructions.
Code: Repository link and embed.
How I Built It: Gemma open-weight AI, AIService abstraction, Ollama local inference, Next.js 14, Monaco editor, Python execution sandbox, and authentic progress data engine.

Why Does Open Innovation Matter?: Student privacy, local inference, eliminating token paywalls, and open pedagogical control.

My Agent Session: Agentic workflow highlights and automated test validation.

Prize Categories: Build for a Friend and Open-Source AI / Gemma Challenge.

MLH second week challage

Sai Rathod — Tue, 06 Oct 2026 17:50:25 +0000

Hello i am build the Trash to Treasure is an offline AI-powered cleanup coach that helps users identify and properly dispose of waste using their phone or laptop camera

What I Built:

Explains the core loop: Go Outside to Spot Litter to Snap Real Photo to Local AI Classifies to Pick Up Safely to Log Real Impact to Pocket Phone & Keep Walking.
Highlights the dedicated Screen-Minimization Mode ("Touch Grass Mode") designed specifically to keep users off their screens and present in nature.
Highlights who it's for (hikers, ploggers, community cleanup groups, students, and eco-explorers).
Demo:

Step-by-step user walkthrough from opening the camera to real-time on-device classification, confirmation checkbox, and confetti impact summary.
Direct links to your live web app and GitHub repository.
Code & Architecture:

Details the Two-Layer Decoupling Architecture:
Layer 1 (Perception): Local Gemma 4 / Ollama vision model observing visual traits.
Layer 2 (Deterministic Rules): Local safety rules and municipal disposal profiles (preventing LLMs from hallucinating city recycling rules).

Highlights the strict zero-state initialization and cryptographic scan_id anti-duplicate point award mechanisms.
Why Does Open Innovation Matter?:

Zero Connectivity Required: Closed APIs fail in state parks and backcountry trails with 0 cellular bars.
100% On-Device Privacy: No personal photos or outdoor locations sent to corporate servers.

Zero Marginal Cost: Local inference on Ollama costs $0.00, removing financial barriers for volunteers.
Auditable Safety: Deterministic open rules prevent dangerous pickups (needles, batteries).

Prize Categories:

Use Gemma, Google's open-weight model, in building your project

GitHub Repository**: GitHub - Trash-to-Treasure

Building an AI Auto-Patch Agent with TrueForge and Qodo

Sai Rathod — Sat, 29 Aug 2026 12:25:12 +0000

Building Auto-Patch Agent: From Production Incident to Reviewed Code Fix

Production incidents require investigation, debugging, code changes, review, and deployment. I built Auto-Patch Agent to automate this workflow while keeping the code change reviewable.

What does Auto-Patch Agent do?

Auto-Patch Agent is an AI-powered incident response agent.

Given a production failure such as an HTTP 500 error, the agent:

Investigates the repository
Finds the root cause
Proposes a targeted fix
Creates a GitHub Pull Request
Gets the change reviewed by Qodo
Merges the fix after review

The Demo

I created a vulnerable Node.js order service for the demonstration.

An invalid customer ID caused the customer lookup to return undefined. The application then tried to access customer.id, resulting in an HTTP 500 error.

The agent investigated the repository and identified the failure path.

It then added validation and improved the error handling so an invalid customer request is handled properly instead of causing an unexpected server error.

The change was committed to a separate branch and submitted as a Pull Request.

How I Used TrueForge

TrueForge is the runtime and orchestration layer for the agent.

I used TrueForge to:

Run the AI agent
Connect Gemini as the model
Connect development and repository tools
Handle agent tool calls
Give the agent access to the project for investigation

This makes the project more than an LLM generating code. The agent interacts with the development workflow.

How I Used Qodo

Qodo provides an independent code review layer.

After the fix was submitted as a Pull Request, Qodo reviewed the changes for bugs and code quality issues.

The review helped validate the remediation before merging.

The final review reported 0 bugs found, and the Pull Request was merged into main.

Workflow

Incident
↓
Repository Investigation
↓
Root Cause
↓
Targeted Fix
↓
GitHub Pull Request
↓
Qodo Review
↓
0 Bugs Found
↓
Merge

Tech Stack

TrueForge
Gemini
GitHub
GitHub MCP
Qodo
Node.js
Express.js

Why I Built It

The goal was to connect an AI model to a real engineering workflow.

Instead of only asking AI for a code suggestion, Auto-Patch Agent investigates the repository, creates a targeted remediation, submits it through GitHub, and gets the change independently reviewed by Qodo.

The final merge remains controlled and reviewable.

This project shows how AI agents can assist with practical incident response and software maintenance.