DEV Community: Manisha

🌿 GrassMapper: Your AI-Powered Guide to Discovering Places in New Cities

Manisha — Sun, 11 Oct 2026 15:41:28 +0000

An open-source AI project built for the Hacktoberfest Touch Grass Challenge.

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

🌱 What I Built

Have you ever travelled to a new city or town and wondered where you could go for a peaceful walk, explore nearby green spaces, or discover interesting places?

When we visit unfamiliar locations, finding places to explore can sometimes be difficult. We often switch between different apps and websites just to decide where to go.

That's why I built GrassMapper — an AI-powered outdoor discovery application designed to help people explore the world around them. 🌍

GrassMapper allows users to discover nearby places based on their location and a selected search radius. It combines location-based discovery with AI-generated recommendations to help users decide where they might want to explore.

✨ Key Features
📍 Nearby Place Discovery: Find places around your current location.
🗺️ OpenStreetMap Integration: Retrieve nearby place information using OpenStreetMap data.
🤖 AI-Powered Recommendations: Generate recommendations using a locally running AI model.
🌳 Travel-Friendly Exploration: Discover places when visiting unfamiliar cities and towns.
📏 Custom Search Radius: Explore places within a selected distance.

My goal is to make discovering new places easier and encourage people to step away from their screens and experience the world around them.

💻 Code

The source code for GrassMapper is available on GitHub.

GitHub Repository: https://github.com/Manisha7530/GrassMapper

The repository contains the project code and instructions for running the application locally.

Make sure this repository exists, is public, and contains your latest code before publishing this post.

🛠️ How I Built It

I built GrassMapper using Python and Flask, combining open-source technologies for location-based discovery and local AI inference.

Technology Stack
Python: Core application logic.
Flask: Backend and web application.
OpenStreetMap: Open geographic data for discovering nearby places.
Overpass API: Querying relevant place information from OpenStreetMap.
Ollama: Running the AI model locally.
Llama 3.2: Generating AI-powered recommendations.
HTML, CSS, and JavaScript: Building the interactive user interface.
How It Works

The application follows a straightforward workflow:

The user selects their current location and search radius.
GrassMapper queries OpenStreetMap data through the Overpass API.
The application displays the retrieved nearby places.
The local AI model generates recommendations based on the available place information and the user's preferences.

One important design choice was using local AI inference through Ollama rather than relying on a hosted LLM API for recommendations.

The application still requires an internet connection to retrieve place data, but the AI inference runs locally.

💚 Why Does Open Innovation Matter?

Open innovation makes it possible for developers to combine existing technologies and build useful solutions without having to create every component from scratch.

For GrassMapper, open-source tools made several things possible.

First, accessible AI development. Using Ollama and an open-weight model allowed me to experiment with local AI inference without depending on a paid, hosted LLM API.

Second, open geographic data. OpenStreetMap provided a way to retrieve nearby place information and connect the application with real-world locations.

Third, flexibility and learning. Working with these tools helped me understand how AI models, APIs, backend systems, and frontend interfaces can work together in a practical application.

I believe open innovation is valuable because it enables developers to experiment, learn, collaborate, and build technology that addresses real-world problems.

GrassMapper is my attempt to apply these principles to something simple but meaningful: helping people discover places and spend more time outdoors. 🌿

🤖 My Agent Session

Optional section.

If you have a DevRelay agent-session link, add it here. Otherwise, you can remove this section.

🏆 Prize Categories

Remove this section if you aren't entering any specific partner prize categories. If you are, list only the categories for which GrassMapper genuinely qualifies.

I Built SquadHero to Help My Friend Stay Consistent With Fitness & Exam Prep

Manisha — Mon, 05 Oct 2026 06:53:48 +0000

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

🛡️ SquadHero: An Accountability Hub Built for My Friend

The Problem

My friend and I have similar goals — staying consistent with fitness, preparing for competitive exams, drinking enough water, and simply not losing motivation after a few days.

The problem is that normal productivity apps are designed for individuals.

And messaging apps are full of distractions.

We needed something simpler:

A shared space where two friends could keep each other accountable.

That idea became SquadHero.

💡 What I Built

SquadHero is a web-based accountability hub designed around three everyday problems:

🏋️ Staying consistent with fitness
📚 Maintaining study/exam preparation routines
🥗 Planning meals safely with roommate allergies

Instead of treating productivity as a boring checklist, SquadHero turns it into a small "hero headquarters" for your squad.

🦸 What Can SquadHero Do?

🔥 1. Shared Accountability

Friends can maintain shared daily goals such as:

Morning workout
Solving practice questions
Drinking enough water
Study sessions
Meal preparation

The interface keeps track of progress and streaks so that accountability becomes visible.

🥗 2. Allergy-Aware Meal Planning

One of the problems I wanted to solve was roommate meal planning.

Different people can have different food allergies.

SquadHero allows roommate allergy profiles and checks recipes against those profiles.

For example:

🚨 Peanut conflict detected for Alex

Instead of simply saying that a recipe is unsafe, the application can suggest alternatives such as replacing peanut butter with sunflower seed butter.

🤖 3. J.A.R.V.I.S. Coach

The project also includes a J.A.R.V.I.S.-style coach for fitness and study planning.

It can work with information such as:

BMI
BMR
TDEE
Water-intake targets
Fitness goals
Exam preparation goals

The goal is to turn a collection of calculations and routines into something that feels like a personal mission-control system.

🕷️ 4. Superhero Reminders

Normal reminders are easy to ignore.

So instead of a boring:

"Drink water."

SquadHero can turn reminders into superhero-style interactions.

The project uses browser-native Web Audio and Speech APIs for interactive audio and voice effects without requiring external audio files.

🛠️ Tech Stack

HTML
CSS
JavaScript
Tailwind CSS
Web Audio API
Web Speech API
LocalStorage
Browser-based animations

The application is intentionally lightweight and can run locally without a complicated build system.

🚀 Running the Project

Clone the repository:

git clone https://github.com/Manisha7530/squadhero-avengers-sync.git
cd squadhero-avengers-sync

Then run:

npm start

Open:

http://localhost:3000