DEV Community: Manan

ClientVault

Manan — Sat, 17 Oct 2026 19:32:16 +0000

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

What I Built

ClientVault is an offline-first tool that strips sensitive client data from contracts and invoices to generate public case studies and tax ledgers. It helps freelancers under strict NDAs safely market their work and organize finances without leaking proprietary client info to the cloud.

Demo

Live App: https://clientvault.onrender.com
Local run: git clone https://github.com/mananmeemroth/ClientVault.git && pip install -r requirements.txt && python app.py

Code

Source code licensed under MIT: https://github.com/mananmeemroth/ClientVault

How I Built It

Built with Python and FastAPI/Gradio, powered by Google's open-weight Gemma 2 (2B) model running local inference via Ollama. Custom system prompts handle entity redaction ([CLIENT], [REDACTED]) and structured financial table extraction entirely in local memory.

Why Does Open Innovation Matter?

Uploading NDA-protected documents to closed cloud APIs breaches client confidentiality and carries per-token costs. Running Gemma locally guarantees zero data leakage outside the device and costs nothing per run.

My Agent Session

Used an AI pair-programmer to craft the entity-redaction prompts for Gemma and generate the turnkey render.yaml configuration for one-click deployment.

My First User

A freelance software consultant tested it on a confidential fintech SOW to draft an anonymized portfolio case study. He confirmed he would pay a one-time $20 fee for an offline tool that protects him from NDA breach disputes.

Prize Categories

Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle
Best Use of Gemma
Best Use of Render

WildNotes - Wonder that grows with you

Manan — Sat, 10 Oct 2026 13:53:34 +0000

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

What I Built

Most modern nature applications suffer from a quiet paradox: they claim to connect us with the outdoors, yet they reduce the living world to a sterile Latin binomial classification on a backlit screen. You snap a photo of a leaf, scan an encyclopedic summary, and immediately slip your phone back into your pocket without ever truly observing the specimen.

WildNotes is the antidote to sterile screen fatigue.

It is an offline-capable, privacy-first field naturalist companion designed specifically to pull you off the screen and into direct sensory contact with the earth.

Instead of delivering academic essays, WildNotes issues The Touch Grass Challenge whenever you frame an outdoor curiosity (moss, lichen, bark, stone, or fungi):

🔎 Look Closer: Guides your eyes toward micro-patterns — fractal vein networks, dewdrop refraction, or spore colony arrangements.
✋ Touch & Feel: Prompts you to physically touch the specimen — testing the damp velvet resilience of moss, furrowed oak bark, or cool river stone.
👃 Scent & Surroundings: Encourages breathing in the microclimate aromas — damp cedar petrichor, crushed pine needles, and companion soil organisms.
🧘 Mindful Pause & Field Inquiry: Posits a grounding question about the specimen's role in the biome and invites you to take three slow, deep breaths outdoors.

Who Is It For?

Mindful Hikers & Trail Walkers seeking more situational presence and less phone scrolling.
Urban Foragers & Backyard Naturalists exploring the hidden micro-habitats growing in sidewalk cracks, brickwork, and neighborhood trees.
Nature Educators & Curious Explorers who believe genuine field observation should engage all five senses.

Demo

🌐 Live Cloud Deployment: https://wildnotes.onrender.com
🔗 GitHub Repository: https://github.com/mananmeemroth/WildNotes

Quick Local Launch


bash
git clone [https://github.com/mananmeemroth/WildNotes.git](https://github.com/mananmeemroth/WildNotes.git)
cd WildNotes
pip install -r requirements.txt
python app.py
# Access the application locally at [http://127.0.0.1:7860](http://127.0.0.1:7860)

Panic2Pass: Turn Exam Panic Into a Plan

Manan — Sun, 04 Oct 2026 20:13:57 +0000

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

What I Built

I built Panic2Pass for my college roommate, who was preparing for an Operating Systems exam with 200+ pages of notes and only a few hours left.

He didn't need another generic chatbot. He needed to know what to study first, what to skip, and how to quickly understand the topics he was struggling with.

Panic2Pass turns his notes and syllabus into four focused tools:

⚡ 30-Min Crash Plan — creates a prioritized revision plan based on the time available.
🆘 I'm Lost — explains difficult concepts in simple language with examples and analogies.
📝 Quiz Me — generates high-value questions from the study material.
📋 Cheat Sheet — turns long notes into a short revision sheet.

I built it specifically around the situation my friend was actually in: "My exam is tomorrow. What should I study right now?"

Demo

🌐 Live Demo: https://panic2pass.onrender.com

The app lets you upload your notes and use the different emergency study modes directly.

Code

The complete project is open source on GitHub:

mananmeemroth / Panic2Pass

🚨 Panic2Pass - The Pre-Exam Emergency AI Cramming Engine

Turn chaotic PDFs, lecture slides, and confusing syllabi into high-yield 30-minute exam triage plans in seconds.

🌟 Overview

Panic2Pass is built for students facing last-minute exam pressure. It takes your uploaded syllabus or lecture notes and synthesizes laser-focused exam rescue materials.

🎯 Emergency Rescue Modes

⚡ 30-Min Crash Plan: Strict triage breakdown (10 mins for formulas/definitions, 15 mins for core mechanisms, 5 mins for hall-door memory checklist).
🆘 I'm Lost (ELI5 Concept Clarifier): De-jargonizes complex topics with intuitive real-world analogies, step-by-step logic, and common misconceptions.
📝 Quiz Me (High-Yield Active Recall): Simulates 3 high-probability exam questions with scoring criteria and hidden model answers.
📋 Rapid Cheat Sheet: Dense markdown matrix of definitions, algorithm steps, and trap checklists.

🚀 Quickstart (Local Development)

1. Ensure Ollama is Running

ollama serve
ollama pull llama3.2:3b

2. Install Dependencies & Run

pip

…

View on GitHub

How I Built It

Panic2Pass is built around an open-weight AI model.

The application takes the student's notes or syllabus and uses the model to generate explanations, quizzes, crash revision plans, and cheat sheets.

For local usage, the project can run with Ollama, allowing the AI to run locally instead of sending study material to a closed AI API.

For the deployed version, the application is hosted on Render with the AI inference configured separately.

The goal was to keep the AI layer flexible so the model can be swapped or run locally without changing the core application.

Why Does Open Innovation Matter?

Study notes can contain personal information, and students shouldn't have to send everything to a closed AI service just to get help studying.

Using an open-weight model gives Panic2Pass the option to run locally, keep notes on the user's machine, and experiment with different models.

It also means the project isn't locked to one AI provider. The model can be changed, improved, or self-hosted as the project evolves.

For me, that's the biggest advantage of open AI: I can actually understand and control the technology I'm building with.

My Agent Session

Prize Categories

Best Use of Render — Panic2Pass is deployed and publicly accessible on Render.