DEV Community: Shivam Mavi

[Boost]

Shivam Mavi — Sat, 17 Oct 2026 17:45:18 +0000

Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle Submission 💸

Shivam Mavi

Oct 17

AquaDrop: Turning Daily Hydration into a Smarter Habit 💧

#devchallenge #hf26challenge

3 min read

AquaDrop: Turning Daily Hydration into a Smarter Habit 💧

Shivam Mavi — Sat, 17 Oct 2026 17:44:02 +0000

What I Built

Meet AquaDrop — a hydration reminder app designed to make drinking water regularly easier, simpler, and more consistent.

We often get busy with college, work, coding, gaming, or everyday tasks and forget something as basic as drinking enough water. AquaDrop aims to solve this problem by making hydration reminders accessible through a clean, friendly interface.

Instead of overwhelming users with complicated health dashboards, AquaDrop focuses on helping them build a small but meaningful daily habit.

Key features

💧 Hydration reminders: Get reminders to take a water break throughout the day.
⏰ Reminder controls: Manage your reminders and snooze them when you need a little more time.
🎨 Light, dark, and system themes: Choose an appearance that suits your preferences.
🔔 Notification and sound experience: Make reminders more noticeable.
✨ A friendly interface: A blue-and-white visual identity with a water-drop-inspired design.
🌱 Room to grow: AquaDrop can evolve into a more comprehensive hydration companion with intake tracking, progress analytics, and personalized coaching.

Who is AquaDrop for?

Students, developers, remote workers, gamers, and anyone who spends long hours concentrating on a screen.

The side-hustle opportunity

AquaDrop could grow into a freemium product: essential hydration reminders for free, with optional premium features such as advanced scheduling, detailed progress insights, and personalization.

The goal is to keep the basic habit-building experience accessible while giving users a reason to upgrade if they find additional features valuable.

Demo

💧 Try AquaDrop live:

https://aqua-drop-seven.vercel.app/

Explore the application and experience its hydration-focused interface.

Code

💻 GitHub repository:

https://github.com/shivammraj/AquaDrop

AquaDrop is an ongoing project, and I want to continue improving it through open-source development, user feedback, and practical experimentation.

How I Built It

I approached AquaDrop as a real product rather than just a landing page or a design exercise.

The focus is on making the core experience straightforward: users should be able to access their reminders, adjust their preferences, and interact with the app without unnecessary complexity.

The project currently has a deployed web experience, and my next priorities include improving reminder reliability, making theme preferences consistent across screens, and expanding hydration tracking.

Where open-source AI fits in

My longer-term vision is to introduce an AI hydration coach that can help users build more consistent habits through contextual tips and personalized insights.

I want to explore open-weight models such as Gemma and local inference options where appropriate. This could make it possible to experiment with AI-powered assistance while retaining greater control over model deployment, costs, and data handling.

Transparency matters: I am distinguishing the current application's functionality from planned AI features. I won't claim that an open-weight model is already integrated unless that integration is implemented and tested.

Why Does Open Innovation Matter?

Open innovation makes it easier for individual developers to turn ideas into real products.

For a side project like AquaDrop, open-source tools and open-weight AI models can help reduce experimentation costs, encourage community contributions, and make it easier to understand and customize the technology behind a product.

They also create opportunities to experiment with local inference and privacy-conscious approaches instead of depending entirely on a closed AI API.

For me, the important part is not simply adding AI because it is popular. It is finding places where open technology can make AquaDrop more useful, affordable, and accessible.

I want to build something that people can actually use, learn from their feedback, and improve it over time.

What's Next?

My roadmap for AquaDrop includes:

Better reminders: Make scheduling, snoozing, and notification behavior more reliable.
Hydration tracking: Let users record their water intake and monitor daily progress.
Progress analytics: Introduce daily and weekly summaries.
AI hydration coach: Explore an open-weight model for contextual habit-building guidance.
Freemium features: Evaluate which advanced features users would genuinely pay for.
Community contributions: Make it easier for other developers to help improve the project.

These are planned improvements, not claims that every feature is already available.

Final Thoughts

AquaDrop started with a simple idea: help people remember to take care of themselves while they work toward their goals.

As a developer, I spend a lot of time thinking about code, projects, and building things. I wanted to create something that addresses a small everyday problem and has the potential to grow into a useful product.

Hacktoberfest's Side Hustle challenge is an opportunity to take that idea further — from a working application toward a sustainable, user-focused product.

Build useful things. Keep learning. Stay hydrated. 💧

If you try AquaDrop, I'd love to hear what you think and what you'd improve!

TouchGrass AI — Turn Screen Time into Real-Life Adventures 🌱Please see tit leave your comment

Shivam Mavi — Sat, 10 Oct 2026 18:58:25 +0000

Shivam Mavi

Oct 10

TouchGrass AI — Turn Screen Time into Real-Life Adventures 🌱

#devchallenge #hf26challenge #ai #webdev

2 min read

TouchGrass AI — Turn Screen Time into Real-Life Adventures 🌱

Shivam Mavi — Sat, 10 Oct 2026 18:57:28 +0000

An open-source AI companion that helps people step away from their screens, explore the world around them, and build healthier offline habits.

What I Built

TouchGrass AI is an AI-powered companion designed to turn the intention of spending less time on screens into real-world action.

Instead of simply telling people to reduce their screen time, it helps them discover meaningful offline activities based on their interests, available time, and surroundings.

From taking a short nature walk and discovering nearby places to trying a new hobby, exercising outdoors, or connecting with friends, TouchGrass AI makes offline experiences easier to start.

The core idea is simple: technology should help us experience more of the real world, not keep us trapped inside a screen.

The project is designed for students, developers, remote workers, and anyone who wants to build a healthier relationship with technology.

Demo

🚀 Live Demo: [https://trail-mate-weld.vercel.app/]

Code

💻 GitHub Repository: [https://github.com/shivammraj/TrailMate]

The project is built with an open-source mindset, with the goal of making AI-powered offline activity discovery accessible, transparent, and customizable.

How I Built It

TouchGrass AI combines an AI-powered activity recommendation experience with a simple, user-friendly interface.

The application is designed around a practical workflow:

Understand the user: Learn about their interests, available time, and preferred activities.
Generate ideas: Use an AI model to suggest realistic, personalized offline activities.
Take action: Help users choose an activity they can complete away from their screens.
Build habits: Encourage users to turn small offline actions into consistent routines.

The project can be extended with location-aware recommendations, outdoor challenges, activity tracking, and privacy-friendly progress summaries.

Technology stack: [https://github.com/shivammraj/TrailMate.]

Open-source AI model: [model I actually used, such as Gemma, Llama, or another open-weight model.]

Rather than treating AI as the destination, the project uses AI as a bridge between digital assistance and real-world experiences.

Why Does Open Innovation Matter?

Open innovation matters because healthier technology should not be limited to people who can afford expensive APIs or proprietary AI services.

Open-weight models give developers more control over how AI systems are deployed, customized, and evaluated. Depending on the model's license and hardware requirements, they can also make local inference possible and reduce dependence on external services.

For a project like TouchGrass AI, these possibilities matter because personal preferences, routines, and location-related information deserve careful handling.

An open approach makes it easier for developers to experiment with privacy-conscious features, adapt recommendations for different communities, and contribute improvements that others can reuse.

The broader vision is to make AI a tool that empowers people to live beyond their devices.

Prize Categories

[1. Gemma

Render
SerpApi]

Final Thoughts

Building with AI is exciting, but the best technology does not always ask for more of our attention.

Sometimes, the best thing an AI can do is help us close the laptop, step outside, and experience something real.

Less scrolling. More living. 🌱

[Boost]

Shivam Mavi — Sun, 04 Oct 2026 08:26:21 +0000

Hacktoberfest Weekend Challenge: Build for a Friend Submission 🤝

Shivam Mavi

Oct 4

CodeMate: I Built a Coding Mentor for My Friend Who Keeps Getting Stuck

#devchallenge #weekendchallenge #hf26challenge

3 min read

CodeMate: I Built a Coding Mentor for My Friend Who Keeps Getting Stuck

Shivam Mavi — Sun, 04 Oct 2026 08:24:00 +0000

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

What I Built

My friend is learning C++ and DSA, but there was a recurring problem: whenever he got stuck on a problem, the easiest solution was to ask an AI for the answer.

That solved the problem once, but it didn't necessarily teach him how to solve the next one.

So I built CodeMate, a local AI coding mentor designed around one simple rule:

Don't give me the answer. Help me figure it out.

CodeMate can:

Give progressive hints instead of immediately revealing solutions
Explain C++ code in beginner-friendly language
Identify bugs and explain why they happen
Analyze a student's approach
Generate practice questions around weak concepts
Adapt the difficulty based on previous attempts

The goal isn't to replace practice platforms. It's to create a learning layer between "I'm stuck" and "I understand it now."

Demo

[https://shivammraj.github.io/CodeMate/]

Code

[https://github.com/shivammraj/CodeMate]

How I Built It

CodeMate is built around a locally running open-weight AI model rather than a proprietary AI API.

The architecture is:

React Frontend
      ↓
FastAPI Backend
      ↓
CodeMate Learning Engine
      ↓
Ollama
      ↓
Open-weight LLM
      ↓
Personalized response

The AI has different modes for different situations:

Hint Mode

The model is instructed not to immediately reveal the complete solution. It gives one conceptual hint and waits for the learner to try again.

Debug Mode

Instead of rewriting the entire program, it identifies the error, explains the cause, and suggests the smallest useful correction.

Explain Mode

It breaks code into its purpose, variables, logic, example execution, and complexity.

Practice Mode

It generates progressively harder problems based on concepts the learner has struggled with.

Why Does Open Innovation Matter?

For this project, using an open model wasn't just a technical choice.

CodeMate is designed around a person's code, mistakes, learning history, and study material. I wanted the core AI experience to be capable of running locally rather than requiring every interaction to be sent to a proprietary cloud API.

Running an open-weight model locally gives CodeMate several important properties:

Privacy: code and learning data can remain on the user's device.
Offline capability: the core AI can work without an internet connection after the model is installed.
Model freedom: the underlying model can be replaced without rebuilding the entire product.
Behavioral control: the tutoring behavior can be changed through our own prompts and application logic.
No per-request API cost: local inference doesn't require paying a proprietary provider for every question.

Open innovation made the AI itself something I could inspect, adapt, and build around rather than simply consume through an API.

What My Friend Thought

After the first version was ready, I gave it to my friend and asked them to use it without explaining how it worked.

Their feedback was:

"[Honestly, CodeMate was helpful because it didn't immediately give me the solution. The hints made me think about the problem myself, which helped me understand the logic better.

I especially liked the Debug Mode because it explained what was wrong with my code instead of just giving me the corrected code.

One thing I would improve is the hint system. Sometimes the first hint was a little too vague, so having a “Give me a stronger hint” option would make it easier when I'm completely stuck.]"

That feedback changed

[Overall, I would actually use it while practicing DSA because it feels more like having someone guide me rather than just giving me the answer.].

What I Learned

The biggest lesson wasn't about getting an AI model to answer coding questions.

It was about building for a specific person.

A generic coding assistant tries to help everyone. CodeMate started with one person, one problem, and one question:

What would actually help my friend learn instead of simply finish the problem?

What's Next

If I continue developing CodeMate, I'd like to add:

Better long-term learning profiles
More accurate misconception detection
Voice-based tutoring
More efficient models for low-end laptops
Personalized revision plans
Support for additional programming languages