DEV Community: jagdish pandey

Ofrenda: Ghost in the Machine - An Offline Altar that works without Internet (Gonda, India)

jagdish pandey — Sat, 31 Oct 2026 16:59:43 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine

What I Built

Ofrenda - Ghost in the Machine: Honoring ancestors with AI that works even in Gonda without internet. 100% OFFLINE.

In Día de los Muertos, families build an ofrenda - an altar with marigolds, candles, the food someone loved. I built a digital one that runs locally.

Live Demo: https://jagdish805212-max.github.io/ofrenda/
Repo: https://github.com/jagdish805212-max/ofrenda

Demo

You answer 7 questions about someone you miss
Local AI (llama3.2:3b via Ollama) writes a remembrance on your own laptop
It builds one HTML file with candles, marigolds, papel picado

Works with WiFi OFF! Tested in Gonda, UP.

How I Built It

Model: llama3.2:3b via Ollama localhost
Frontend: Single HTML file, vanilla JS
Storage: localStorage only, no upload
No cloud, no API key

Why Ghost in the Machine?

Ghost is not AI hallucination. Ghost is our ancestors' stories. Machine just helps keep them.

Built with love from Gonda, Uttar Pradesh, India.

LazyMate AI — Work Less, Live More: Your Open-Source AI Productivity Assistant

jagdish pandey — Sat, 24 Oct 2026 05:55:41 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady

🦥 LazyMate AI — Work Less, Live More

What if your productivity assistant could help you get things done without sending all your notes to a paid AI service?

Meet LazyMate AI, a lightweight productivity assistant designed to reduce repetitive work, organize messy notes, and turn overwhelming tasks into small, manageable steps.

The idea is simple: spend less time organizing work and more time actually doing what matters.

What I Built

LazyMate AI is a productivity website built with HTML, CSS, and JavaScript, with optional integration for local open-weight AI models through Ollama.

It helps users:

✅ Manage tasks: Add, complete, and delete daily tasks.
🧠 Summarize notes: Turn long notes into concise key points.
🪜 Create action plans: Break goals into practical, step-by-step tasks.
🎯 Prioritize work: Focus on what matters instead of feeling overwhelmed.
💾 Save tasks locally: Keep task data in browser storage.
🔒 Explore private AI: Process notes through a locally running model instead of relying on a hosted AI API.

The goal is to make productivity feel lighter, especially for students and people juggling multiple responsibilities.

Demo

https://github.com/jagdish805212-max/lazymate-ai.git

Live demo: [Add your deployed website link here]

Code

GitHub repository: file:///C:/Users/Acer/Desktop/LAZY%20%20MIT/index.html

The project uses three core files:

LazyMate-AI/
├── index.html
├── style.css
├── script.js
└── README.md

No frontend framework is required. The website uses standard web technologies, making it easy for beginners to explore, modify, and contribute to.

How I Built It

I built LazyMate AI using:

HTML5 for the website structure.
CSS3 for the responsive dashboard and visual design.
JavaScript for task management, local persistence, and productivity helpers.
Ollama for optional local inference with an installed open-weight model, such as Gemma.

The application has two modes.

1. Demo mode

The core task manager works locally in the browser. Basic summaries and action plans can also be generated using JavaScript rules without a model server.

2. Local AI mode

When Ollama is running and correctly configured, LazyMate can send prompts to a locally installed model for more flexible summaries and action plans.

This distinction matters: rule-based fallback results are not AI-generated. Real model responses require a working local inference setup.

Why Does Open Innovation Matter?

For me, open innovation is about giving people more control over the tools they use.

A productivity assistant may handle personal notes, study plans, assignments, and unfinished tasks. I wanted to explore an approach where users can choose to process that information on their own computer.

Using an open-weight model locally offers several advantages:

More control over data: Notes can be processed locally without sending them to a third-party AI API.
Model flexibility: Developers can experiment with different compatible models.
Lower recurring API costs: Local inference does not require paying a provider for every request, although hardware and electricity still cost money.
Room to experiment: Prompts and application behavior can be customized.
Offline potential: Once the model is installed and configured, local inference can work without an internet connection.

These benefits depend on the actual configuration. The optional local AI mode requires a compatible model runtime, and the initial model download requires internet access.

I like that open tools let developers start small, learn by building, and improve a project without depending entirely on a proprietary AI service.

My Agent Session

I built the project through an iterative development process, starting with the frontend and adding task management, local storage, and the optional AI integration.

Agent session: [Add your DevRelay session link here, if available]

The Lazy Payoff 🦥

Imagine having a page of messy class notes and not knowing where to start.

Instead of manually reorganizing everything, you can ask LazyMate to identify the main points and suggest a small next step.

The project aims to reduce the friction between having a task and knowing how to begin it.

I have not measured a verified time saving yet, so I won't claim a specific number. The next step is to test the app with real study notes and everyday tasks, then measure how much time it saves.

Sometimes, being productive doesn't mean working harder. It means making the next step easier.

Prize Categories

Primary theme: Slow and Steady — Open-Source AI.

I'll select any additional partner categories only if the project meets their specific requirements.

Final Thoughts

Building LazyMate AI helped me explore how ordinary web technologies can work with open-weight AI to create something practical.

It's a small project, but the idea behind it is bigger: AI should give people more control, make repetitive work easier, and help them conserve their energy for things that matter.

Work less. Live more. Stay a little slothful. 🌱🦥

Hacktoberfest #OpenSource #AI #JavaScript #Ollama #BuildInPublic

HustleMate AI-Turn Your Skills into Side Hustle with AI

jagdish pandey — Wed, 14 Oct 2026 07:43:33 +0000

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

HustleMate AI is an AI-powered side-hustle assistant for students and beginners. It helps users turn their skills like coding, designing, writing into real earning opportunities. It suggests what service to offer, how to find clients, and gives a complete roadmap to start freelancing.

Demo

Live Demo: https://jagdish805212-max.github.io/hustlemate-ai/

Code

GitHub Repo: https://github.com/jagdish805212-max/hustlemate-ai

How I Built It

Built with HTML, CSS, JavaScript and deployed on GitHub Pages.

Why Does Open Innovation Matter?

Open innovation helps beginners learn faster. By using open-source tools, I was able to build HustleMate AI without high costs and share it with everyone for free.

My Agent Session

My First User

Prize Categories

StudyBuddy AI – Learn Smarter, Live Better

jagdish pandey — Fri, 09 Oct 2026 03:55:07 +0000

StudyBuddy AI — AI Assistant for Your Friend 🌱🤖

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

What I Built

I built StudyBuddy AI, an AI-powered study companion designed to make learning easier, more interactive, and more enjoyable.

Students often spend hours in front of screens while studying, which can make learning tiring and reduce their time for real-world activities. StudyBuddy AI aims to make studying more engaging while encouraging students to take breaks, step away from their screens, and connect with the world around them.

The goal is simple: Study smarter, take healthy breaks, and learn together!

StudyBuddy AI is designed for students who want a friendly digital companion to support their learning journey.

Key Goals

🤖 Make studying more interactive with AI.
📚 Help students learn and understand concepts.
🌱 Encourage screen breaks and offline activities.
🤝 Make learning feel more friendly and less stressful.

Demo

GitHub Repository:
https://github.com/jagdish805212-max/studybuddy-ai

Live Demo: Coming soon.

Note: The current website runs locally on my computer. I'll share a public demo link after deploying it.

Code

Explore the source code and contribute to the project on GitHub:

👉 https://github.com/jagdish805212-max/studybuddy-ai

The project is an opportunity to explore how web technologies and AI can be combined to create a student-friendly learning experience.

How I Built It

I built StudyBuddy AI as a web-based project using HTML, CSS, and JavaScript for the user interface and interactive features.

The project focuses on creating a simple, accessible, and user-friendly experience for students.

My development approach includes:

HTML: Structuring the website.
CSS: Designing the layout and visual appearance.
JavaScript: Adding interactivity and application logic.
AI integration: Supporting the study-assistant experience.

My goal is to explore open-source AI and build a useful tool that supports learning without encouraging students to remain glued to their screens.

AI model and framework details will be added according to the actual model and tools integrated into the project.

Why Does Open Innovation Matter?

Open innovation makes it possible for developers, students, and communities to learn from one another and build better solutions together.

For a project like StudyBuddy AI, open-source development can help me:

Learn from other developers and their implementations.
Improve the project through community feedback.
Make the code accessible for contributions and collaboration.
Experiment with open AI technologies and understand how they work.
Build tools that can be adapted to different learning needs.

Open innovation is especially important for student developers because it provides an opportunity to learn by building real projects and collaborating with the global developer community.

I believe AI should not only make us more productive online but also help us build healthier habits in the real world.

My Agent Session

I don't have a public agent session to share yet.

I'll add a DevRelay session link here if I record and publish one.

Prize Categories

Primary challenge: Hacktoberfest Open-Source AI Challenge — Week 1: Touch Grass.

I'll add any additional partner prize categories if the project meets their specific requirements.

Final Thoughts

StudyBuddy AI is my step toward building AI-powered tools that make learning more helpful, friendly, and balanced.

As a student developer, I'm excited to keep learning, improve the project, and welcome feedback from the open-source community.

Let's use AI to learn better, connect more, and touch grass! 🌱🚀

GitHub: https://github.com/jagdish805212-max/studybuddy-ai