Draku: Exorcising the Ghosts in Your Camera Roll with Offline Open-Weight AI 🦇 Hacktoberfest Open-Source AI Challenge: Ghost in the Machine Submission 👻🌼 Devansh Follow Oct 31 #devchallenge…

Description: "A 100% offline, privacy-first personal AI vault preserving family oral histories, memories, and personal anecdotes. Built for Hacktoberfest 2026: Ghost in the Machine." tags: "devchallenge, hacktoberfest, opensource, ai" canonical_url: false cover_image: _This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine_ > _"Some voices deserve to live forever."_ ## What I Built Every family has stories that live only in the voices of aging grandparents, faded handwritten index cards, or casual Sunday afternoon conversations: * The exact way Nani brewed cardamom chai when the monsoon started. * The afternoon Uncle Leo held his trenchcoat over a drenched saxophone player during the 1989 Jazz Festival in the rain. * Grandpa Joe's 1968 blue Chevy pickup and the summer road trips before interstates existed. When these voices fade, those stories are often lost forever. **Echoes** is an offline, privacy-first personal AI application engineered to preserve family oral histories, memories, and personal anecdotes using **100% open-source AI running directly on your own computer**. ### Interpreting the Theme: _Ghost in the Machine_ In folklore and fiction, a "ghost" is an echo of a life once lived. While technology often tries to simulate deceased people with conversational bots, we believe that crossing into supernatural imitation is ethically fraught and emotionally dissonant. We interpreted the theme through a grounded, human lens: 1. **The Machine** : A 100% offline local computer running open-weight language models and speech recognition engines in complete isolation from the cloud. 2. **The Ghost** : The indelible human spirit captured in authentic voice recordings, spoken memories, and family narratives. 3. **The Creative Track ("Ghost Story Mode")** : A playful, vintage CRT terminal where users can select keepsake props from real memories (an old watch, a recipe, a jazz cassette) and have the local AI weave them into atmospheric gothic campfire tales—always prominently certified as 100% artistic fiction. ### Key Capabilities * **🏛️ The Memory Vault** : Record voice stories in your browser or upload audio files (`.wav`, `.mp3`, `.m4a`), paired with rich text narratives, dates, relatives, and tags. Search everything instantly with offline hybrid search. * **🎙️ Local Voice Transcription** : Converts spoken recordings to text locally via Whisper—no audio ever leaves your computer. * **📻 AI Memory Assistant (RAG)** : Chat with your family memories. The assistant answers questions strictly using your saved stories, citing exact memories as sources and refusing to invent or hallucinate facts. * **🕯️ Remembrance Studio** : Create a digital tribute honoring a loved one—weaving their favorite songs, comfort foods, life motto, and memories into a respectful memorial message with a standalone printable HTML keepsake. * **👻 Ghost Story Mode** : Retro green CRT scanline terminal with a 1-to-5 Horror Intensity Slider (_Cozy Twilight_ to _Gothic Nightmare_) that spins creative campfire fiction inspired by keepsake motifs. * **🛡️ 100% Zero-Cloud Privacy Guarantee** : No cloud API keys. No OpenAI, Gemini, or Claude. No telemetry. Your memories belong to your family alone. ## Demo ### Video Walkthrough An interactive animated walkthrough demonstrating the Memory Vault, Local Transcription, RAG Memory Assistant, Remembrance Studio, and Ghost Story Terminal: _(A high-resolution demo recording is included in the repository at`docs/media/echoes_app_tour_1793469836415.webp`)_ ### Live Interface Screenshots Memory Vault & Detail Modal | Remembrance Studio Tribute ---|--- | Ghost Story Mode CRT Terminal | Voice Keepsake Recorder ---|--- | ### Local Quickstart (Ready in Under 2 Minutes) Because Echoes is built for total privacy, it runs entirely on your local machine: bash # 1. Clone repository git clone https://github.com/ashutosh-ranjan/ghost-in-the-machine.git cd ghost-in-the-machine # 2. Start local AI model (using Ollama) ollama pull llama3.2:3b # 3. Launch both backend & frontend # On Windows PowerShell: .\scripts\run_all.ps1 # Or with Docker Compose: docker compose up --build

How a Fake Font File Almost Hacked My GitHub Kaushal Verma Kaushal Verma Kaushal Verma Follow Oct 27 ## How a Fake Font File Almost Hacked My GitHub #security #devchallenge 1 reaction Add Comment 1 min read

Hello DEV Community! 👋 As a beginner participating in Hacktoberfest 2026, I am exploring how open-source AI models and tools are shaping modern development. During Week 3, I learned about open-weight AI models and how developer communities collaborate around open-source machine learning projects. I am excited to continue my journey in open source! # hacktoberfest #devchallenge

LedgerPrep Lite: Turning Receipt Chaos into Bookkeeping-Ready Records with Local AI devchallenge hf26challenge ai opensource Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle Submission 💸 This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle What I Built Imagine running a small business and keeping receipts in WhatsApp, invoices in email, and other bills in your phone gallery. At the end of the month, someone has to collect everything, enter the details, find missing information, and prepare the records for the accountant. That is the problem LedgerPrep Lite aims to solve. LedgerPrep Lite is a local AI-powered bookkeeping preparation tool designed for small business owners and independent bookkeepers. Users upload invoice and receipt images, and the app extracts important information such as the seller's name, invoice number, date, and total amount. It also flags possible duplicate invoices, missing information, and totals that may need checking. Users can review and correct the results before exporting an organized CSV for their existing accounting workflow. The goal isn't to replace accountants. It's to reduce the manual work that happens before the actual accounting begins. Demo https://zenkaicode123-leaderprep-app-3xbjxn.streamlit.app/] The demo shows the journey from uploading scattered receipts to reviewing extracted information and exporting organized records. Code https://github.com/ZenkaiCode123/ledgerprep-.git How I Built It I built LedgerPrep Lite with a simple Python-based stack, keeping the application focused on one complete workflow instead of building an entire accounting platform. Streamlit: Provides the user interface for uploading documents, reviewing results, and downloading records. Ollama and Gemma 3 4B: Run an open-weight language model locally to extract structured information from supported invoice and receipt images. Python: Validates extracted data and flags missing fields, potential duplicates, and arithmetic inconsistencies where sufficient information is available. Pandas: Organizes the reviewed records and exports them to CSV. The model's output is not automatically treated as correct. Financial information can be ambiguous, so users must be able to inspect the original document, correct mistakes, and review flagged records. Why Does Open Innovation Matter? Financial documents contain sensitive business information. Sending every receipt and invoice to an external AI service isn't always desirable for a small business. By using an open-weight model through local inference, LedgerPrep Lite is designed to process documents on the user's own computer without requiring a hosted AI API for extraction. Open innovation also gives us more control over the technology. We can experiment with different models, change the extraction process, and evaluate their performance without being locked into a single provider's API pricing. For this project, open-source AI isn't just a technical choice. It supports the goal of making AI-assisted bookkeeping preparation more private and affordable. Prize Categories Best Use of Gemma

LedgerVault AI — Private, Zero-Cloud Invoice & Receipt Sorter for Freelancers Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle Submission 💸 Atharva Lawange Atharva Lawange Atharva Lawange Follow Oct 14 ## LedgerVault AI — Private, Zero-Cloud Invoice & Receipt Sorter for Freelancers #devchallenge #hf26challenge 1 reaction 1 comment 4 min read

Keep reading to learn more about Hacktoberfest and meet Mochi the Explorer! 🩷

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass ## What I Built **OutsideQuest 🌿** is an outdoor adventure buddy that helps people spend less time scrolling and more time exploring the world around them. Users choose how much time they have and what interests them, and OutsideQuest provides a practical outdoor mission with clear steps, an estimated duration, and safety guidance. For example, a 30-minute break could become a gentle walk along a familiar path, noticing three details in nature. It's designed for people who want to get outside without needing elaborate plans, expensive equipment, or an entire free day. ## Demo 🌿 **Live demo:** https://outsidequest.onrender.com/ The public demo is available directly in your browser, with no API keys or local setup required. ## Code 💻 **GitHub repository:** https://github.com/shrey04kaitke-sys/OutsideQuest The project is built with Python and Streamlit, with a local AI-powered mission-generation workflow and fallback missions. ## How I Built It I built OutsideQuest using: * **Python** for application logic. * **Streamlit** for the interactive web interface. * **Ollama + Qwen2.5 3B** for local, open-weight AI mission generation. * **Safety validation and fallback missions** to make the experience more practical and reliable. * **Render** for public deployment. The local version uses Ollama and Qwen2.5 3B to generate missions. The deployed public demo uses built-in missions, so visitors can try the application without needing to run a local model or configure API keys. I also designed the public demo to disable personal progress and reflection storage, keeping those features available for local use instead. ## Why Does Open Innovation Matter? Open innovation gave me the flexibility to build an AI experience around my own requirements rather than relying entirely on a closed, hosted API. Using an open-weight model with local inference lets users experiment with mission generation on their own machines, with greater control over the model and processing environment. It also makes the project easier to customize, extend, and learn from. I can experiment with prompts, improve safety validation, and explore different models without making a proprietary API the foundation of the entire experience. For me, open source is about more than making code publicly available. It's about giving other developers the freedom to understand, adapt, and build upon an idea. ## Prize Categories * **Best Use of Render** — OutsideQuest is deployed on Render, making the public demo accessible through a browser without requiring users to install the application.

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

# devchallenge #weekendchallenge #hf26challenge Hacktoberfest Weekend Challenge: Build for a Friend 🤝** **This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend** **What I Built** I built ScamCheck for a friend who regularly receives suspicious bank, KYC, delivery, payment and account-warning messages and usually asks me: GitHub: github.com/ansh-rajpoot/scamcheck "Is this real or a scam?" The problem is that simply asking an AI chatbot isn't enough. A chatbot can confidently say "this looks like a scam", but that doesn't tell you why, what can actually be verified, or what is still unknown. So I built something different. ScamCheck investigates the message and shows the evidence. Paste a suspicious SMS, WhatsApp message, email, or URL and ScamCheck breaks it into: suspicious signals it directly observed organizations and claims mentioned in the message URLs and domains external evidence found through web search things that could not be verified safe next steps The AI doesn't get to invent evidence or blindly decide the answer. How it works _**DEMO**_ * The result looks more like a security investigation than a chatbot: * Suspicious signals found * Observed: "Your account will be blocked today" * Observed: OTP requested * Claimed organization: SBI * External evidence: relevant official/public source * Unverified: sender identity * Recommended action: don't share the OTP; verify through the organization's official website/app instead. There is deliberately no arbitrary "87% scam" score. Evidence should be explainable, not a number invented by an LLM. Why Local AI? Scam messages can contain extremely sensitive information: ** 1. - bank details 2. - account suffixes 3. - phone numbers 4. - transaction information 5. - names 6. - addresses 7. - payment information ** ScamCheck uses **Gemma 3 4B **through Ollama, so the AI analysis can run locally on the user's machine. The entire private message doesn't need to be sent to an AI API just to understand it. When external verification is useful, ScamCheck sends a minimal search query rather than the complete private message. That's the part of open-weight AI that mattered to me: the user can actually keep the core analysis at home. _**Why not just use an LLM? Because an LLM shouldn't be trusted with facts it can determine more reliably using normal code. **_ For example: "Enter your OTP immediately" can be detected deterministically. So can: https://example.com/login Gemma is then used for the parts where language understanding is useful: extracting claims identifying claimed organizations understanding requested actions structuring the investigation External search provides another layer of evidence. The final report keeps those sources separate. The Friend Test The real test isn't whether ScamCheck produces a clever AI response. It's whether my friend can paste a suspicious message and understand: "What exactly is suspicious here, what was actually verified, and what should I do next?" That's what I built it for. ## Tech Stack *_Python FastAPI HTML/CSS/JavaScript Ollama Gemma 3 4B SerpApi *_ The application runs locally, with the browser communicating with a local FastAPI server. ## Open Source AI The core AI runs through an open-weight model rather than requiring a proprietary hosted model for the analysis. That means the model can be: run locally swapped for another open model inspected and controlled by the developer used without sending every private message to a third-party AI API For a tool dealing with potentially sensitive messages, that difference matters. What I Learned The biggest lesson was that AI should not be the entire security system. The useful architecture turned out to be: deterministic code + local AI + external evidence rather than: send everything to an LLM and trust its verdict. That made ScamCheck both more explainable and more useful. Limitations ScamCheck cannot guarantee that a message is legitimate or fraudulent. External search results are evidence, not absolute proof. It also does not automatically access WhatsApp, Gmail, Telegram, or other private accounts, and it does not automatically open suspicious URLs. The goal is to help a person investigate safely—not pretend that an AI can know everything. Prize Categories Gemma — Best Use of Gemma SerpApi — Best Use of SerpApi

This is a submission for the Sanity Challenge, Path Two: Vibe-Code Something Strange At 16:50, a...

Sanity Challenge — Path Two: Vibe-Code Something Strange At 16:50, a photograph is supposed to...

Congrats to the Frontend Challenge: Comfort Food Edition Winners! Pure CSS cat bakery wins top spot Jem Jem Jem Follow for The DEV Team Sep 10 ## Congrats to the Frontend Challenge: Comfort Food Edition Winners! #devchallenge #frontendchallenge #css #javascript 47 reactions 21 comments 2 min read

Contributing to one of my personal Bookmarks manager - and making it better for everyone Summer Bug Smash: Clear the Lineup 🐛🛹 Abdul Samad Abdul Samad Abdul Samad Follow Aug 27 ## Contributing to one of my personal Bookmarks manager - and making it better for everyone #bugsmash #devchallenge #opensource #ai 5 reactions Add Comment 9 min read

Making a dog talk with zero backend: in-browser segmentation, WebGL lip sync and real MP4 export DEV Weekend Challenge: Dog Days Edition Submission 🐕 Amandeep Singh Amandeep Singh Amandeep Singh Follow Aug 16 ## Making a dog talk with zero backend: in-browser segmentation, WebGL lip sync and real MP4 export #devchallenge #weekendchallenge #ai 6 reactions Add Comment 6 min read

This is a submission for the Notion MCP Challenge What I Built OpenClaw just got an...