DEV Community: Ishlok Dwivedi

WebBloom : Idea into Website

Ishlok Dwivedi — Sat, 24 Oct 2026 10:28:17 +0000

I Built OfferLint WebBloom: Turn Your Idea Into a Website

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

What I Built

I’m working on OfferLint, a project that helps people turn their website ideas into a website preview.

The idea is pretty simple. Many small business owners want a website, but they don't always know how to build one or want to spend a lot of money getting started.

With OfferLint, users can describe what kind of website they need, choose a design style, and generate a preview with the help of AI. They can check how their website might look before deciding what to do next.

I’m also trying out a simple pricing model:

₹199 one-time: Get the website source code and setup instructions.
₹499/month: A planned OfferLint+ subscription for deployment help, maintenance, and support.

Right now, I'm focusing on getting the basic experience right and making it useful for small businesses, freelancers, and people starting their own projects.

Demo

I'm currently testing OfferLint locally and working towards a public deployment.

Live demo: Add my Render URL here after deployment.
Code: Add my GitHub repository link here.
Screenshots/video: I'll include a short demo showing how a user enters an idea and gets a website preview.

Code

GitHub: Add your public repository link here.

The project uses Python and FastAPI for the backend, along with HTML, CSS, and JavaScript for the frontend.

How I Built It

I wanted to make the process straightforward instead of building another chatbot.

The user describes their website idea, selects the features and design they want, and gets a website preview. I'm using predefined website layouts along with AI-generated content to make the result more relevant to each business.

The main technologies I'm using are:

Python and FastAPI
HTML, CSS, and JavaScript
Gemini API for AI-generated content
Razorpay integration for the planned checkout flow

I'm still improving the project, especially the generated designs and the complete purchase experience.

Why Does Open Innovation Matter?

Open-source tools made it easier for me to build and experiment without having to create everything from scratch.

I can work on the backend, change the website templates, test different AI models, and let other developers contribute improvements.

I also want to explore open-weight models in the future so I can compare different approaches and have more control over how the project works.

For me, the main point is to make website creation more accessible to people who might not otherwise be able to afford a custom website.

What's Next?

There are still a few things I want to work on:

Improve the quality of generated website designs.
Make the downloaded source code more complete and useful.
Deploy the project and test it properly.
Finish testing payments and the purchase flow.
Get feedback from people who actually need a website.

I'm treating this as an early-stage project, so there is still a lot to learn and improve.

Prize Categories

I'm entering the general Side Hustle challenge with OfferLint.

I'll only add a partner-specific category if my final implementation meets its requirements.

Thanks for checking out my project! I'd love to hear your suggestions on what I should improve next.

OfferLint: Turn Your Idea Into a Website

Ishlok Dwivedi — Tue, 13 Oct 2026 08:22:57 +0000

I Built OfferLint: Turn Your Idea Into a Website

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

What I Built

I’m working on OfferLint, a project that helps people turn their website ideas into a website preview.

The idea is pretty simple. Many small business owners want a website, but they don't always know how to build one or want to spend a lot of money getting started.

I’m also trying out a simple pricing model:

₹199 one-time: Get the website source code and setup instructions.
₹499/month: A planned OfferLint+ subscription for deployment help, maintenance, and support.

Right now, I'm focusing on getting the basic experience right and making it useful for small businesses, freelancers, and people starting their own projects.

How I Built It

I wanted to make the process straightforward instead of building another chatbot.

The main technologies I'm using are:

Python and FastAPI
HTML, CSS, and JavaScript
Gemini API for AI-generated content
Razorpay integration for the planned checkout flow

I'm still improving the project, especially the generated designs and the complete purchase experience.

Why Does Open Innovation Matter?

Open-source tools made it easier for me to build and experiment without having to create everything from scratch.

I can work on the backend, change the website templates, test different AI models, and let other developers contribute improvements.

I also want to explore open-weight models in the future so I can compare different approaches and have more control over how the project works.

For me, the main point is to make website creation more accessible to people who might not otherwise be able to afford a custom website.

What's Next?

There are still a few things I want to work on:

Improve the quality of generated website designs.
Make the downloaded source code more complete and useful.
Deploy the project and test it properly.
Finish testing payments and the purchase flow.
Get feedback from people who actually need a website.

I'm treating this as an early-stage project, so there is still a lot to learn and improve.

Prize Categories

I'm entering the general Side Hustle challenge with OfferLint.

I'll only add a partner-specific category if my final implementation meets its requirements.

Thanks for checking out my project! I'd love to hear your suggestions on what I should improve next.

AeroGemma: Touch Grass with AI-Powered Air Quality (Powered by Google Gemma)

Ishlok Dwivedi — Tue, 06 Oct 2026 15:54:23 +0000

AeroGemma: Checking If You Can Actually Touch Grass Today 🌿

I spend most of my day in front of a screen, so when the Touch Grass challenge came up, I thought it fit me pretty well. But I also noticed a problem. Going outside isn't always a good idea. If the air is full of PM2.5, ozone and smog, a run or a walk can do more harm than good. I'm from Delhi NCR, so I know this one personally.

So I built AeroGemma, a web app that tells you whether it's safe to go outside right now. It pulls live air quality data and then uses Google's Gemma model (gemma-4-26b-a4b-it) to turn the numbers into advice a normal person can follow.

Demo
GitHub Repo: IshlokDwivedi-coder/AQI_System
Run it locally: node server.js, or just open index.html in your browser.
What it does
🌿 "Can I Touch Grass Today?" verdict: a color gauge and a badge that tell you if the air is okay for going out.
🏃 Activity checker: separate status for Running & Cardio, Gardening, Dog Walking & Hikes, and Kids' Playgrounds.
🤖 Gemma Outdoor Advisor: Gemma reads the pollutant values and gives you simple advice on what to do.
📍 Search any city: use GPS in one click or search by city name. It uses Open-Meteo, so there's no API key or rate limit to worry about.
🛡️ Offline mode: if there's no internet, a built-in rule engine still gives you a basic answer.

It also breaks down 6 pollutants (PM2.5, PM10, O3, NO2, SO2, CO) and compares them against WHO guideline values.

Code

The whole project is open source:

IshlokDwivedi-coder / AQI_System

🌿 AeroGemma — Touch Grass with AI-Powered Air Quality

Submission for Hacktoberfest Open-Source AI Challenge: Week 1 ("Touch Grass")
Prize Category Track: 🏆 Best Use of Gemma (Google's Open-Weight Model)
Live GitHub Repo: IshlokDwivedi-coder/AQI_System
Author: Ishlok Dwivedi (@ishlokdwivedicoder)

📖 Overview

The theme of this week's Hacktoberfest challenge is Touch Grass: build something with open-source AI at its core that gets people away from screens and out into the real world.

While spending time in nature (hiking, running, gardening, birding) is vital for physical and mental health, poor air quality (elevated PM2.5, ground-level ozone, and chemical smog) can turn a refreshing outdoor workout into a health hazard—especially for children, older adults, and those with respiratory conditions.

AeroGemma solves this dilemma. It is a sleek, intuitive, and responsive web application that fetches real-time atmospheric readings from global monitoring stations and pairs them with Google's open-weight Gemma model (gemma-4-26b-a4b-it…

View on GitHub

Repo URL: https://github.com/IshlokDwivedi-coder/AQI_System

How I Built It

I kept the stack simple on purpose.

Frontend: plain HTML, CSS and JavaScript, with no framework. The page theme changes color with the AQI level using CSS variables, and the gauge is an SVG. I also tried a glassmorphism look, which was fun to figure out.
Gemma: I call Gemma through the Google AI Studio API. The tricky part was prompt design. I send it the live values (US AQI, PM2.5, PM10, O3, NO2, temperature and UV index) and ask it for practical health advice. It took a few tries before the answers stopped sounding generic.
Air quality data: Open-Meteo's Air Quality and Weather API, which covers stations worldwide and doesn't need a token.
Offline fallback: if the user has no signal, a simple rule-based engine gives a basic recommendation so the app still does something useful.
Why I Used an Open Model

I picked Gemma instead of a closed API for a few reasons that actually matter for an app like this:

It can work without internet. Hiking trails and national parks often have no signal, and a cloud-only API is useless there. Open weights can run on a laptop or phone.
Privacy. To use a closed API you have to send your location and health details (like asthma) to someone else's servers. With an open model, that data can stay on your device.
No cost, no lock-in. College clubs, run groups and NGOs can build on it without paying per token or worrying that an API will be shut down.

Honestly, I think this is where open models matter most: small, useful tools that don't need a big company behind them.

My Agent Session

I built and deployed this with pair-programming help from DevRelay:

Building AeroGemma: Touch Grass AQI Advisor with Google Gemma

You

I want to build an Air Quality Index (AQI) website for the Hacktoberfest 'Touch Grass' challenge using Google's open-weight Gemma model, and push it to GitHub.

Agent

Great concept! We will build AeroGemma: a real-time AQI dashboard connected to Open-Meteo station data and powered by Google Gemma (gemma-4-26b-a4b-it) to evaluate whether it is safe to touch grass and recommend tailored outdoor activities.

▸ write_to_file index.html, styles.css, app.js, server.js, README.md

Input

index.html, styles.css, app.js, server.js, README.md

Output

Created modern responsive UI, dynamic AQI status theme shifts, Open-Meteo station integration, and Google Gemma prompt engine.

You

Make sure all code is in AQI_System so I can push it directly to my GitHub repo.

Agent

Committed and pushed clean code to https://github.com/IshlokDwivedi-coder/AQI_System with GitHub Secret Scanning compliance.

▸ run_command git push -u origin main

Input

$ git push -u origin main

Output

Branch 'main' set up to track 'origin/main'.

(Session link: https://dev.to/agent_sessions/building-aerogemma-touch-grass-aqi-advisor-with-google-gemma-pld3yg)

I built Saarthi, a voice assistant that phones my friend Krishna's grandmother, Malti Devi and for SBI customer's

Ishlok Dwivedi — Fri, 02 Oct 2026 04:28:26 +0000

This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend

What I Built

I built Saarthi for my friend Krishna Singh's grandmother, Malti Devi and SBI cutomer's . Talk about only Malti Devi here she lives alone, and she doesn't use a touchscreen phone. Most apps today assume you can tap, type and swipe, so for someone like her, solving even a small problem online is almost impossible.

Saarthi fixes that with just her voice. The AI agent actually calls her. If she wants to talk or ask something, she just speaks, and the agent understands her, finds the answer and talks back. There is no typing, no screen, no menus.

Demo

Code

IshlokDwivedi-coder / Saarthi

PARAKH — परख

parakh (परख): "to assess, to gauge the true quality of something."

A transparent, deterministic, network-free candidate ranker for the Redrob × Hack2Skill Intelligent Candidate Discovery & Ranking challenge. It ranks the top 100 candidates out of 100,000 for the Senior AI Engineer — Founding Team role using an explainable panel of feature-engineered scorers — no embeddings, no ML model weights, and zero network calls during ranking.

Every score is fully decomposable and every line of the reasoning column is assembled only from real fields on the candidate's profile, so hallucination is impossible by construction.

TL;DR — reproduce the submission

Single command (local, requires JDK 17):

./gradlew bootJar && java -jar build/libs/parakh.jar <path-to>/candidates.jsonl ./submission.csv

Single command (Docker — no JDK needed, runs with the network physically off):

# 1. put the contest file at ./data/candidates.jsonl
mkdir -p data && cp <path-to>/candidates.jsonl ./data/

…

View on GitHub

How I Built It

The backend is written in Java with Spring Boot. I used the Gemini API for the AI part: it converts her speech to text, works out what she is asking, and generates the reply. The reply then goes back to her as speech, so she hears the answer instead of reading it.

The interface works in both modes, text and speak, so anyone can use it either way.

Why Does Open Innovation Matter?

Most technology is built for people who can use a touchscreen. People like Malti Devi are left out, and they are often the ones who are alone and need help the most. Saarthi shows that a simple voice call can give them access to the same answers everyone else gets.

What Krishna Said

When I told Krishna about it, he was shocked. He was really happy, and it was a very emotional and enjoyable moment for him.

Thanks for giving me this opportunity