<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Aishwarya</title>
    <description>The latest articles on DEV Community by Aishwarya (@aishwarya_15).</description>
    <link>https://dev.to/aishwarya_15</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4156075%2F1c71f132-5f7e-42ef-9d13-b72f1e6508fe.png</url>
      <title>DEV Community: Aishwarya</title>
      <link>https://dev.to/aishwarya_15</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/aishwarya_15"/>
    <language>en</language>
    <item>
      <title>👻 WRAITH</title>
      <dc:creator>Aishwarya</dc:creator>
      <pubDate>Sat, 31 Oct 2026 13:05:04 +0000</pubDate>
      <link>https://dev.to/aishwarya_15/wraith-4c0m</link>
      <guid>https://dev.to/aishwarya_15/wraith-4c0m</guid>
      <description>&lt;p&gt;&lt;em&gt;This is a submission for the &lt;a href="https://dev.to/challenges/hacktoberfest-week4-2026-10-26"&gt;Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine&lt;/a&gt;&lt;/em&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  What I Built
&lt;/h2&gt;

&lt;p&gt;Every developer knows the feeling.&lt;/p&gt;

&lt;p&gt;Your website works perfectly during development. Every button responds, every form submits, and every page loads.&lt;/p&gt;

&lt;p&gt;Then someone clicks something you never expected.&lt;/p&gt;

&lt;p&gt;And suddenly, something breaks.&lt;/p&gt;

&lt;p&gt;Traditional automated tests follow predefined instructions. But what about unexpected user journeys, unusual interface states, or bugs hiding between interactions?&lt;/p&gt;

&lt;p&gt;That's the problem WRAITH aims to solve.&lt;/p&gt;

&lt;p&gt;WRAITH is an AI-assisted website testing platform designed to investigate applications like a digital paranormal investigator.&lt;/p&gt;

&lt;p&gt;Instead of relying exclusively on manually scripted tests, its proposed AI agent will plan browser interactions, explore authorized websites, observe unexpected behavior, and generate evidence-backed defect reports.&lt;/p&gt;

&lt;p&gt;The Ghost-Hunting Toolkit&lt;/p&gt;

&lt;p&gt;🦇 Command Center — Enter the Haunted Domain&lt;/p&gt;

&lt;p&gt;A centralized dashboard for initiating investigations, reviewing activity, and monitoring testing progress.&lt;/p&gt;

&lt;p&gt;🕯️ Investigations — Follow the Digital Footprints&lt;/p&gt;

&lt;p&gt;An investigation history designed to organize testing sessions, target websites, and discovered issues.&lt;/p&gt;

&lt;p&gt;⚰️ Bug Graveyard — Where Broken Code Goes to Rest&lt;/p&gt;

&lt;p&gt;A centralized defect registry featuring severity-based filtering, issue tracking, and editable resolution status.&lt;/p&gt;

&lt;p&gt;👻 Ghost Replay — Revisit the Scene&lt;/p&gt;

&lt;p&gt;A planned mechanism for reproducing failures through recorded browser actions and supporting evidence.&lt;/p&gt;

&lt;p&gt;📜 Reports — The Evidence Never Lies&lt;/p&gt;

&lt;p&gt;Structured investigation summaries and downloadable findings.&lt;/p&gt;

&lt;p&gt;🔥 Exorcism Mode — Make Sure the Ghost Is Gone&lt;/p&gt;

&lt;p&gt;A planned regression-testing capability that will rerun relevant checks after developers implement fixes.&lt;/p&gt;

&lt;p&gt;Why the Halloween Theme?&lt;/p&gt;

&lt;p&gt;The challenge is called Ghost in the Machine, and I wanted to interpret that theme through something developers encounter every day.&lt;/p&gt;

&lt;p&gt;Software bugs are digital ghosts.&lt;/p&gt;

&lt;p&gt;They appear unexpectedly, disappear when someone tries to reproduce them, and return when everyone thinks the problem has been solved.&lt;/p&gt;

&lt;p&gt;WRAITH turns that frustrating experience into a ghost-hunting workflow.&lt;/p&gt;

&lt;p&gt;The Halloween-inspired black-and-orange interface isn't just decoration. It gives each testing capability a memorable identity while preserving the structure of a practical developer tool.&lt;/p&gt;

&lt;h2&gt;
  
  
  Code
&lt;/h2&gt;


&lt;div class="ltag-github-readme-tag"&gt;
  &lt;div class="readme-overview"&gt;
    &lt;h2&gt;
      &lt;img src="https://assets.dev.to/assets/github-logo-5a155e1f9a670af7944dd5e12375bc76ed542ea80224905ecaf878b9157cdefc.svg" alt="GitHub logo"&gt;
      &lt;a href="https://github.com/aishwarya150624" rel="noopener noreferrer"&gt;
        aishwarya150624
      &lt;/a&gt; / &lt;a href="https://github.com/aishwarya150624/Wraith" rel="noopener noreferrer"&gt;
        Wraith
      &lt;/a&gt;
    &lt;/h2&gt;
    &lt;h3&gt;
      This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine.  Some bugs hide in plain sight. Others haunt your application long after you think they're gone.  What if your website had a ghost hunter?  Meet WRAITH, an open-source AI website testing agent concept designed to explore web application.
    &lt;/h3&gt;
  &lt;/div&gt;
&lt;/div&gt;


&lt;h2&gt;
  
  
  How I Built It
&lt;/h2&gt;

&lt;p&gt;I started with a question:&lt;/p&gt;

&lt;p&gt;What would automated website testing look like if the testing tool could decide what to investigate next?&lt;/p&gt;

&lt;p&gt;That question became the foundation of WRAITH.&lt;/p&gt;

&lt;p&gt;Phase 1: Designing the Investigation Experience&lt;/p&gt;

&lt;p&gt;I first mapped the complete testing workflow:&lt;/p&gt;

&lt;p&gt;Target Website → Testing Objective → AI Planning → Browser Exploration → Anomaly Detection → Evidence Collection → Report → Replay&lt;/p&gt;

&lt;p&gt;Instead of building a conventional testing dashboard, I designed an interface around the idea of a paranormal investigation.&lt;/p&gt;

&lt;p&gt;Every screen represents a stage of discovering, documenting, and resolving software defects.&lt;/p&gt;

&lt;p&gt;The result is Black Ember, a functional frontend prototype that demonstrates the intended user experience.&lt;/p&gt;

&lt;p&gt;Phase 2: Planning the Open-Source AI Architecture&lt;/p&gt;

&lt;p&gt;WRAITH's proposed architecture combines several open-source technologies.&lt;/p&gt;

&lt;p&gt;🧠 Qwen + Ollama — The Reasoning Engine&lt;/p&gt;

&lt;p&gt;The planned AI component uses open-weight Qwen models through Ollama for local inference.&lt;/p&gt;

&lt;p&gt;The agent will interpret testing objectives, propose browser actions, and evaluate observations.&lt;/p&gt;

&lt;p&gt;🎭 Playwright — The Browser Investigator&lt;/p&gt;

&lt;p&gt;Playwright will execute browser interactions such as navigation, clicking buttons, completing forms, and inspecting application behavior.&lt;/p&gt;

&lt;p&gt;It will also support evidence collection through screenshots and browser traces.&lt;/p&gt;

&lt;p&gt;🔗 LangGraph — The Agent Workflow&lt;/p&gt;

&lt;p&gt;LangGraph is planned to coordinate the investigation loop:&lt;/p&gt;

&lt;p&gt;Plan → Act → Observe → Evaluate → Continue or Report&lt;/p&gt;

&lt;p&gt;This structure is intended to let the agent adjust its next action based on what happens inside the browser.&lt;/p&gt;

&lt;p&gt;⚡ FastAPI — The Backend&lt;/p&gt;

&lt;p&gt;A Python/FastAPI backend will connect the frontend to browser automation, AI inference, and investigation management.&lt;/p&gt;

&lt;p&gt;Phase 3: Building Toward Autonomous Testing&lt;/p&gt;

&lt;p&gt;The next milestone is turning the simulated investigation workflow into real browser execution.&lt;/p&gt;

&lt;p&gt;The planned implementation will prioritize authorized testing targets, controlled execution, reproducible evidence, and validation of AI-generated findings.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why Does Open Innovation Matter?
&lt;/h2&gt;

&lt;p&gt;For WRAITH, open innovation is more than a development preference.&lt;/p&gt;

&lt;p&gt;It's fundamental to the kind of tool I want to build.&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Developers Should Control Their Testing Tools&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;A website testing agent may encounter sensitive application workflows, internal interfaces, and development environments.&lt;/p&gt;

&lt;p&gt;Local AI inference can help reduce the need to transmit application information to external AI services.&lt;/p&gt;

&lt;p&gt;Using open-weight models through Ollama is intended to make that possible.&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Intelligence Should Be Customizable&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;Different applications require different testing strategies.&lt;/p&gt;

&lt;p&gt;An e-commerce application, student portal, and enterprise dashboard have completely different user journeys.&lt;/p&gt;

&lt;p&gt;An open-source architecture allows developers to customize agent prompts, testing objectives, evaluation rules, and browser workflows.&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Testing Should Be Reproducible&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;A useful bug report needs more than an AI-generated explanation.&lt;/p&gt;

&lt;p&gt;Developers need evidence.&lt;/p&gt;

&lt;p&gt;WRAITH's planned architecture emphasizes browser traces, screenshots, logs, and reproducible action sequences.&lt;/p&gt;

&lt;p&gt;Open tooling makes those mechanisms inspectable and extensible.&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Innovation Should Be Accessible&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;Not every developer or student can afford recurring commercial AI API costs.&lt;/p&gt;

&lt;p&gt;By designing WRAITH around open-weight models and locally executable tooling, I want to make experimentation more accessible.&lt;/p&gt;

&lt;p&gt;The goal isn't simply to build an AI that finds bugs. It's to build an AI testing workflow that developers can understand, inspect, modify, and improve together.&lt;/p&gt;

&lt;h2&gt;
  
  
  Prize Categories
&lt;/h2&gt;

&lt;p&gt;Hacktoberfest Open-Source AI Challenge — Week 4: Ghost in the Machine&lt;/p&gt;

&lt;p&gt;Partner-specific prize categories and eligibility will be finalized based on the implemented integrations and challenge requirements.&lt;/p&gt;

</description>
      <category>devchallenge</category>
      <category>hf26challenge</category>
    </item>
    <item>
      <title>Sloth Digest: Slow news. Taken at your own pace.</title>
      <dc:creator>Aishwarya</dc:creator>
      <pubDate>Sun, 25 Oct 2026 12:17:42 +0000</pubDate>
      <link>https://dev.to/aishwarya_15/sloth-digest-slow-news-taken-at-your-own-pace-3e4c</link>
      <guid>https://dev.to/aishwarya_15/sloth-digest-slow-news-taken-at-your-own-pace-3e4c</guid>
      <description>&lt;p&gt;&lt;em&gt;This is a submission for the &lt;a href="https://dev.to/challenges/hacktoberfest-week3-2026-10-19"&gt;Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady&lt;/a&gt;&lt;/em&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  What I Built
&lt;/h2&gt;

&lt;h2&gt;
  
  
  Demo
&lt;/h2&gt;


&lt;div class="crayons-card c-embed text-styles text-styles--secondary"&gt;
    &lt;div class="c-embed__content"&gt;
      &lt;div class="c-embed__body flex items-center justify-between"&gt;
        &lt;a href="https://accounts.google.com/v3/signin/identifier?continue=https://drive.google.com/file/d/1QbBXPCnHOiPi9ilrQMMTDxFOZhAIMBoi/view?usp%3Dsharing&amp;amp;followup=https://drive.google.com/file/d/1QbBXPCnHOiPi9ilrQMMTDxFOZhAIMBoi/view?usp%3Dsharing&amp;amp;osid=1&amp;amp;passive=1209600&amp;amp;service=wise&amp;amp;flowName=WebLiteSignIn&amp;amp;flowEntry=ServiceLogin&amp;amp;dsh=S-954739755:1792930661726447" rel="noopener noreferrer" class="c-link fw-bold flex items-center"&gt;
          &lt;span class="mr-2"&gt;accounts.google.com&lt;/span&gt;
          

        &lt;/a&gt;
      &lt;/div&gt;
    &lt;/div&gt;
&lt;/div&gt;


&lt;h2&gt;
  
  
  Code
&lt;/h2&gt;


&lt;div class="ltag-github-readme-tag"&gt;
  &lt;div class="readme-overview"&gt;
    &lt;h2&gt;
      &lt;img src="https://assets.dev.to/assets/github-logo-5a155e1f9a670af7944dd5e12375bc76ed542ea80224905ecaf878b9157cdefc.svg" alt="GitHub logo"&gt;
      &lt;a href="https://github.com/aishwarya150624" rel="noopener noreferrer"&gt;
        aishwarya150624
      &lt;/a&gt; / &lt;a href="https://github.com/aishwarya150624/sloth-digest" rel="noopener noreferrer"&gt;
        sloth-digest
      &lt;/a&gt;
    &lt;/h2&gt;
    &lt;h3&gt;
      app to help lazy and boring tasks
    &lt;/h3&gt;
  &lt;/div&gt;
  &lt;div class="ltag-github-body"&gt;
    
&lt;div id="readme" class="md"&gt;&lt;div class="markdown-heading"&gt;
&lt;h1 class="heading-element"&gt;🦥 Sloth Digest&lt;/h1&gt;
&lt;/div&gt;
&lt;p&gt;Wake up to a 2-line summary of your group chats, written by an AI that never leaves your laptop.&lt;/p&gt;
&lt;p&gt;Export a chat (WhatsApp: &lt;em&gt;Chat → More → Export chat → Without media&lt;/em&gt;), drop the &lt;code&gt;.txt&lt;/code&gt; in a folder, and run one command. Sloth Digest reads only the &lt;strong&gt;new&lt;/strong&gt; messages since last time and writes a digest:&lt;/p&gt;
&lt;div class="snippet-clipboard-content notranslate position-relative overflow-auto"&gt;&lt;pre class="notranslate"&gt;&lt;code&gt;## family_group (13 new messages)
TL;DR: &amp;lt;two short sentences&amp;gt;
Plans
- &amp;lt;dates, places&amp;gt;
Needs you
- [ ] &amp;lt;things asked of you&amp;gt;
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;
&lt;div class="markdown-heading"&gt;
&lt;h2 class="heading-element"&gt;Why open-source AI?&lt;/h2&gt;

&lt;/div&gt;
&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Private:&lt;/strong&gt; your family chats never touch a cloud API. Summarizing them with a hosted model means handing a stranger your conversations.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Free forever:&lt;/strong&gt; a nightly run over many chats would add up on a paid API. Locally it costs electricity.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Offline:&lt;/strong&gt; works without internet.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Swappable:&lt;/strong&gt; &lt;code&gt;--model qwen2.5:7b&lt;/code&gt; for sharper summaries, or a small model for an old laptop.&lt;/li&gt;
&lt;/ul&gt;
&lt;div class="markdown-heading"&gt;
&lt;h2 class="heading-element"&gt;Built for laziness&lt;/h2&gt;

&lt;/div&gt;
&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Incremental:&lt;/strong&gt; remembers what it…&lt;/li&gt;
&lt;/ul&gt;&lt;/div&gt;
  &lt;/div&gt;
  &lt;div class="gh-btn-container"&gt;&lt;a class="gh-btn" href="https://github.com/aishwarya150624/sloth-digest" rel="noopener noreferrer"&gt;View on GitHub&lt;/a&gt;&lt;/div&gt;
&lt;/div&gt;


&lt;h2&gt;
  
  
  How I Built It
&lt;/h2&gt;

&lt;h2&gt;
  
  
  🛠️ How I Built It
&lt;/h2&gt;

&lt;p&gt;Building &lt;strong&gt;Sloth Digest&lt;/strong&gt; started with a simple idea: &lt;strong&gt;What if technology helped us slow down instead of constantly demanding our attention?&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;For the Hacktoberfest Week 3: Slow and Steady challenge, I wanted to create something that made information consumption simpler, calmer, and more intentional.&lt;/p&gt;

&lt;h3&gt;
  
  
  🦥 1. Designing a Slower Experience
&lt;/h3&gt;

&lt;p&gt;I began by defining the core philosophy of Sloth Digest: quality over quantity.&lt;/p&gt;

&lt;p&gt;Instead of creating another traditional news feed, I designed the experience around a single weekly issue containing meaningful stories, interesting insights, and positive developments.&lt;/p&gt;

&lt;p&gt;The goal was to eliminate unnecessary browsing and make staying informed feel effortless.&lt;/p&gt;

&lt;h3&gt;
  
  
  📰 2. Structuring the Weekly Digest
&lt;/h3&gt;

&lt;p&gt;Each issue follows a simple reading flow:&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Weekly Snapshot → Curated Stories → Reading Progress&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;The weekly snapshot provides a quick overview, while curated articles encourage readers to explore topics more deeply.&lt;/p&gt;

&lt;p&gt;A gentle progress&lt;/p&gt;

&lt;h2&gt;
  
  
  Why Does Open Innovation Matter?
&lt;/h2&gt;

&lt;h2&gt;
  
  
  🌍 Why Does Open Innovation Matter?
&lt;/h2&gt;

&lt;p&gt;&lt;strong&gt;Because meaningful innovation should be accessible to everyone, not locked behind expensive subscriptions, proprietary systems, or closed ecosystems.&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;While building &lt;strong&gt;Sloth Digest&lt;/strong&gt;, I wanted to explore a different relationship between technology and attention. Instead of creating another platform designed to keep people scrolling, I envisioned something that helps them consume less information but understand more.&lt;/p&gt;

&lt;p&gt;That's where the philosophy of open innovation becomes especially meaningful.&lt;/p&gt;

&lt;h3&gt;
  
  
  🦥 Freedom to Build Differently
&lt;/h3&gt;

&lt;p&gt;Open-source technology gives developers the freedom to experiment, customize, and improve their solutions without being entirely dependent on a single company's infrastructure.&lt;/p&gt;

&lt;p&gt;For a project like Sloth Digest, this flexibility matters.&lt;/p&gt;

</description>
      <category>devchallenge</category>
      <category>hf26challenge</category>
    </item>
    <item>
      <title>AI Can Write Your Proposal. But Can It Prove What It Says? Meet PITCHCRAFT.</title>
      <dc:creator>Aishwarya</dc:creator>
      <pubDate>Sun, 18 Oct 2026 03:27:42 +0000</pubDate>
      <link>https://dev.to/aishwarya_15/ai-can-write-your-proposal-but-can-it-prove-what-it-says-meet-pitchcraft-m5b</link>
      <guid>https://dev.to/aishwarya_15/ai-can-write-your-proposal-but-can-it-prove-what-it-says-meet-pitchcraft-m5b</guid>
      <description>&lt;p&gt;&lt;em&gt;This is a submission for the &lt;a href="https://dev.to/challenges/hacktoberfest-week2-2026-10-12"&gt;Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle&lt;/a&gt;&lt;/em&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  What I Built
&lt;/h2&gt;

&lt;h2&gt;
  
  
  Demo
&lt;/h2&gt;


&lt;div class="crayons-card c-embed text-styles text-styles--secondary"&gt;
    &lt;div class="c-embed__content"&gt;
      &lt;div class="c-embed__body flex items-center justify-between"&gt;
        &lt;a href="https://accounts.google.com/v3/signin/identifier?continue=https://drive.google.com/file/d/1rFa0sM14uYcdwUQ3KcGSNKLAkaBzsL3n/view?usp%3Dsharing&amp;amp;followup=https://drive.google.com/file/d/1rFa0sM14uYcdwUQ3KcGSNKLAkaBzsL3n/view?usp%3Dsharing&amp;amp;osid=1&amp;amp;passive=1209600&amp;amp;service=wise&amp;amp;flowName=WebLiteSignIn&amp;amp;flowEntry=ServiceLogin&amp;amp;dsh=S-852792771:1792294283567841" rel="noopener noreferrer" class="c-link fw-bold flex items-center"&gt;
          &lt;span class="mr-2"&gt;accounts.google.com&lt;/span&gt;
          

        &lt;/a&gt;
      &lt;/div&gt;
    &lt;/div&gt;
&lt;/div&gt;


&lt;h2&gt;
  
  
  Code
&lt;/h2&gt;


&lt;div class="ltag-github-readme-tag"&gt;
  &lt;div class="readme-overview"&gt;
    &lt;h2&gt;
      &lt;img src="https://assets.dev.to/assets/github-logo-5a155e1f9a670af7944dd5e12375bc76ed542ea80224905ecaf878b9157cdefc.svg" alt="GitHub logo"&gt;
      &lt;a href="https://github.com/aishwarya150624" rel="noopener noreferrer"&gt;
        aishwarya150624
      &lt;/a&gt; / &lt;a href="https://github.com/aishwarya150624/Pitchcraft" rel="noopener noreferrer"&gt;
        Pitchcraft
      &lt;/a&gt;
    &lt;/h2&gt;
    &lt;h3&gt;
      PITCHCRAFT — An evidence-first AI proposal studio for freelancers. Create personalized, portfolio-backed proposals, verify claims, and track opportunities.
    &lt;/h3&gt;
  &lt;/div&gt;
  &lt;div class="ltag-github-body"&gt;
    
&lt;div id="readme" class="md"&gt;&lt;div class="markdown-heading"&gt;
&lt;h1 class="heading-element"&gt;PITCHCRAFT&lt;/h1&gt;
&lt;/div&gt;
&lt;p&gt;A local-first proposal workspace for freelancers. Paste a client brief and get a proposal drafted
only from your own profile and portfolio. Every sentence is traced to its source, and anything
unbacked is flagged. No build step, no dependencies, no backend. Data lives in your browser.&lt;/p&gt;
&lt;div class="markdown-heading"&gt;
&lt;h2 class="heading-element"&gt;Run it&lt;/h2&gt;
&lt;/div&gt;
&lt;ol&gt;
&lt;li&gt;Open this folder in VS Code and accept the recommended extensions.&lt;/li&gt;
&lt;li&gt;Right-click &lt;code&gt;index.html&lt;/code&gt; and choose &lt;strong&gt;Open with Live Server&lt;/strong&gt; (port 5500), or run &lt;code&gt;npm start&lt;/code&gt; (port 3000).&lt;/li&gt;
&lt;li&gt;Press F5 to open it in Chrome with the included launch configs.&lt;/li&gt;
&lt;/ol&gt;
&lt;p&gt;Opening &lt;code&gt;index.html&lt;/code&gt; directly from disk also works. The Google Fonts link needs an internet connection
without it the page falls back to Georgia and system fonts.&lt;/p&gt;
&lt;div class="markdown-heading"&gt;
&lt;h2 class="heading-element"&gt;Structure&lt;/h2&gt;

&lt;/div&gt;
&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;File&lt;/th&gt;
&lt;th&gt;Purpose&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;&lt;code&gt;index.html&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;Page shell and script order&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;code&gt;css/styles.css&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;All styling, with light and dark theme tokens&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;code&gt;js/state.js&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;Sample data, localStorage persistence, small helpers&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;code&gt;js/engine.js&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;
&lt;code&gt;check&lt;/code&gt; (claim verification), &lt;code&gt;analyse&lt;/code&gt; (brief to needs, fit, draft), &lt;code&gt;segs&lt;/code&gt;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;…&lt;/div&gt;
  &lt;/div&gt;
  &lt;div class="gh-btn-container"&gt;&lt;a class="gh-btn" href="https://github.com/aishwarya150624/Pitchcraft" rel="noopener noreferrer"&gt;View on GitHub&lt;/a&gt;&lt;/div&gt;
&lt;/div&gt;


&lt;h2&gt;
  
  
  How I Built It
&lt;/h2&gt;

&lt;h2&gt;
  
  
  Why Does Open Innovation Matter?
&lt;/h2&gt;

&lt;h2&gt;
  
  
  Prize Categories
&lt;/h2&gt;

</description>
      <category>devchallenge</category>
      <category>hf26challenge</category>
    </item>
    <item>
      <title>🌿 WANDER — An AI App That Wants You to Go Outside An app that gives you a reason to go outside, then gets out of the way.</title>
      <dc:creator>Aishwarya</dc:creator>
      <pubDate>Sun, 11 Oct 2026 05:49:59 +0000</pubDate>
      <link>https://dev.to/aishwarya_15/wander-an-ai-app-that-wants-you-to-go-outside-an-app-that-gives-you-a-reason-to-go-outside-1h9b</link>
      <guid>https://dev.to/aishwarya_15/wander-an-ai-app-that-wants-you-to-go-outside-an-app-that-gives-you-a-reason-to-go-outside-1h9b</guid>
      <description>&lt;p&gt;&lt;em&gt;This is a submission for the &lt;a href="https://dev.to/challenges/hacktoberfest-week1-2026-10-05"&gt;Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass&lt;/a&gt;&lt;/em&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  What I Built
&lt;/h2&gt;

&lt;h2&gt;
  
  
  🌿 What I Built
&lt;/h2&gt;

&lt;p&gt;&lt;strong&gt;What if an AI app encouraged you to stop using it?&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;Meet &lt;strong&gt;WANDER&lt;/strong&gt; — an offline-first AI companion that turns everyday surroundings into little adventures.&lt;/p&gt;

&lt;p&gt;Powered by &lt;strong&gt;Google's Gemma 3 4B through Ollama&lt;/strong&gt;, WANDER uses local AI to generate three personalized outdoor missions based on your mood, available time, weather, and previous explorations. Gemma helps keep each adventure fresh and relevant without relying on cloud-based AI.&lt;/p&gt;

&lt;p&gt;🌱 &lt;strong&gt;Choose&lt;/strong&gt; your mood and available time.&lt;/p&gt;

&lt;p&gt;✨ &lt;strong&gt;Discover&lt;/strong&gt; three unique missions generated by Gemma 3.&lt;/p&gt;

&lt;p&gt;📵 &lt;strong&gt;Disconnect&lt;/strong&gt; as your screen goes dark.&lt;/p&gt;

&lt;p&gt;🌍 &lt;strong&gt;Explore&lt;/strong&gt; your surroundings and rediscover overlooked details.&lt;/p&gt;

&lt;p&gt;📖 &lt;strong&gt;Reflect&lt;/strong&gt; in your private Wander Journal and track your &lt;strong&gt;Wander Ratio&lt;/strong&gt; — time outside versus time on screen.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;WANDER doesn't measure success by how long you stay, but by how much you experience after leaving. 🌿&lt;/strong&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  Demo
&lt;/h2&gt;


&lt;div class="crayons-card c-embed text-styles text-styles--secondary"&gt;
    &lt;div class="c-embed__content"&gt;
      &lt;div class="c-embed__body flex items-center justify-between"&gt;
        &lt;a href="https://accounts.google.com/v3/signin/identifier?continue=https://drive.google.com/file/d/1C5y5wN8xzWsUKSpAFc_7Kvjq0CjllhSH/view?usp%3Dsharing&amp;amp;followup=https://drive.google.com/file/d/1C5y5wN8xzWsUKSpAFc_7Kvjq0CjllhSH/view?usp%3Dsharing&amp;amp;osid=1&amp;amp;passive=1209600&amp;amp;service=wise&amp;amp;flowName=WebLiteSignIn&amp;amp;flowEntry=ServiceLogin&amp;amp;dsh=S-1087636011:1791697798181576" rel="noopener noreferrer" class="c-link fw-bold flex items-center"&gt;
          &lt;span class="mr-2"&gt;accounts.google.com&lt;/span&gt;
          

        &lt;/a&gt;
      &lt;/div&gt;
    &lt;/div&gt;
&lt;/div&gt;


&lt;h2&gt;
  
  
  Code
&lt;/h2&gt;


&lt;div class="ltag-github-readme-tag"&gt;
  &lt;div class="readme-overview"&gt;
    &lt;h2&gt;
      &lt;img src="https://assets.dev.to/assets/github-logo-5a155e1f9a670af7944dd5e12375bc76ed542ea80224905ecaf878b9157cdefc.svg" alt="GitHub logo"&gt;
      &lt;a href="https://github.com/aishwarya150624" rel="noopener noreferrer"&gt;
        aishwarya150624
      &lt;/a&gt; / &lt;a href="https://github.com/aishwarya150624/WANDER" rel="noopener noreferrer"&gt;
        WANDER
      &lt;/a&gt;
    &lt;/h2&gt;
    &lt;h3&gt;
      🌿 WANDER — An offline-first AI companion powered by Gemma 3 that generates personalized outdoor missions to help you disconnect from screens and rediscover the real world.
    &lt;/h3&gt;
  &lt;/div&gt;
  &lt;div class="ltag-github-body"&gt;
    
&lt;div id="readme" class="md"&gt;&lt;div class="markdown-heading"&gt;
&lt;h1 class="heading-element"&gt;WANDER (Day 1 build)&lt;/h1&gt;
&lt;/div&gt;
&lt;div class="markdown-heading"&gt;
&lt;h2 class="heading-element"&gt;Run&lt;/h2&gt;
&lt;/div&gt;
&lt;ol&gt;
&lt;li&gt;
&lt;code&gt;ollama pull gemma3:4b&lt;/code&gt;  (any model name works: &lt;code&gt;WANDER_MODEL=qwen2.5:7b&lt;/code&gt;)&lt;/li&gt;
&lt;li&gt;&lt;code&gt;pip install -r requirements.txt&lt;/code&gt;&lt;/li&gt;
&lt;li&gt;Optional weather: &lt;code&gt;export WANDER_LAT=.. WANDER_LON=..&lt;/code&gt;
&lt;/li&gt;
&lt;li&gt;&lt;code&gt;uvicorn main:app --host 0.0.0.0 --port 8010&lt;/code&gt;&lt;/li&gt;
&lt;li&gt;On your phone (same Wi-Fi) open &lt;code&gt;http://&amp;lt;laptop-ip&amp;gt;:8010&lt;/code&gt;
&lt;/li&gt;
&lt;/ol&gt;
&lt;div class="markdown-heading"&gt;
&lt;h2 class="heading-element"&gt;What exists&lt;/h2&gt;
&lt;/div&gt;
&lt;ul&gt;
&lt;li&gt;
&lt;code&gt;/api/tasks&lt;/code&gt;: 3 tasks, schema-enforced, random ingredient, history-aware, safety check, repeat filter&lt;/li&gt;
&lt;li&gt;
&lt;code&gt;/api/finish&lt;/code&gt;, &lt;code&gt;/api/journal&lt;/code&gt;: SQLite journal with wander ratio&lt;/li&gt;
&lt;li&gt;
&lt;code&gt;safety_log&lt;/code&gt; table keeps every rejected task, for the "what the checker caught" count&lt;/li&gt;
&lt;/ul&gt;
&lt;div class="markdown-heading"&gt;
&lt;h2 class="heading-element"&gt;Known gaps (Day 2)&lt;/h2&gt;

&lt;/div&gt;
&lt;ul&gt;
&lt;li&gt;Photo + caption, season layer, request log / privacy receipt, eval scripts&lt;/li&gt;
&lt;li&gt;Screen time counts from "Get three tasks" to saving, not the setup screen before it&lt;/li&gt;
&lt;li&gt;Not yet run against a real model: the first real run will show if prompts need tuning&lt;/li&gt;
&lt;/ul&gt;
&lt;div class="markdown-heading"&gt;
&lt;h2 class="heading-element"&gt;Day 2 additions&lt;/h2&gt;

&lt;/div&gt;
&lt;ul&gt;
&lt;li&gt;Photo on the return screen: resized on the phone, captioned by the local vision model, never judged&lt;/li&gt;
&lt;li&gt;Season layer: prompt hint by date (autumn: leaf colour stages…&lt;/li&gt;
&lt;/ul&gt;&lt;/div&gt;
  &lt;/div&gt;
  &lt;div class="gh-btn-container"&gt;&lt;a class="gh-btn" href="https://github.com/aishwarya150624/WANDER" rel="noopener noreferrer"&gt;View on GitHub&lt;/a&gt;&lt;/div&gt;
&lt;/div&gt;


&lt;h2&gt;
  
  
  How I Built It
&lt;/h2&gt;

&lt;h2&gt;
  
  
  🛠️ How I Built It
&lt;/h2&gt;

&lt;p&gt;I designed WANDER around a simple idea: &lt;strong&gt;AI should encourage exploration, not endless interaction.&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;The project uses an offline-first architecture with the following technologies:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;&lt;p&gt;*&lt;em&gt;Gemma 3 4B *&lt;/em&gt; Runs AI locally to generate personalized outdoor missions based on mood, time, weather, and previous activities.&lt;/p&gt;&lt;/li&gt;
&lt;li&gt;&lt;p&gt;&lt;strong&gt;SQLite:&lt;/strong&gt; Stores mission history and private journal entries locally.&lt;/p&gt;&lt;/li&gt;
&lt;li&gt;&lt;p&gt;&lt;strong&gt;HTML, CSS &amp;amp; JavaScript:&lt;/strong&gt; Power the lightweight, mobile-friendly interface&lt;/p&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The AI generates three missions, checks them for safety and repetition, and lets users choose one before putting their phones away.&lt;/p&gt;

&lt;p&gt;I also designed a &lt;strong&gt;Wander Ratio&lt;/strong&gt; to measure time spent outdoors versus time on screen, alongside a &lt;strong&gt;Privacy Receipt&lt;/strong&gt; feature to make data usage transparent.&lt;/p&gt;

&lt;p&gt;The biggest focus was keeping WANDER &lt;strong&gt;private, lightweight, and genuinely useful beyond the screen.&lt;/strong&gt; 🌿&lt;/p&gt;

&lt;h2&gt;
  
  
  Why Does Open Innovation Matter?
&lt;/h2&gt;

&lt;h2&gt;
  
  
  🌍 Why Does Open Innovation Matter?
&lt;/h2&gt;

&lt;p&gt;I believe &lt;strong&gt;technology should give us more freedom, not demand more of our attention.&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;Open innovation is what makes WANDER possible. By using &lt;strong&gt;Gemma 3 through Ollama&lt;/strong&gt;, WANDER can generate personalized outdoor missions locally without depending on expensive cloud APIs.&lt;/p&gt;

&lt;p&gt;Here's why it matters:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;🔒 &lt;strong&gt;Privacy First:&lt;/strong&gt; Personal reflections, photos, and activity history can stay on the user's device.&lt;/li&gt;
&lt;li&gt;🌐 &lt;strong&gt;Offline Freedom:&lt;/strong&gt; Core AI features can work without an internet connection.&lt;/li&gt;
&lt;li&gt;🔄 &lt;strong&gt;Flexibility:&lt;/strong&gt; Developers can experiment with different open-weight models and customize the experience.&lt;/li&gt;
&lt;li&gt;💚 &lt;strong&gt;Accessibility:&lt;/strong&gt; No per-request API fees, making the project easier to explore and improve.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;A closed API wouldn't offer the same level of offline independence, privacy, and control.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;For me, open innovation isn't just about sharing code. It's about building AI that respects people's privacy, choices, and time.&lt;/strong&gt; 🌿&lt;/p&gt;

&lt;h2&gt;
  
  
  Prize Categories
&lt;/h2&gt;

&lt;h3&gt;
  
  
  Best Use of Gemma 
&lt;/h3&gt;

&lt;p&gt;WANDER uses &lt;strong&gt;Google's Gemma 3 4B open-weight model&lt;/strong&gt;, running locally through Ollama, to generate personalized outdoor missions based on users' moods, available time, surroundings, and previous activities.&lt;/p&gt;

&lt;p&gt;Gemma is central to WANDER's goal of making AI-powered exploration private, offline-first, and accessible without relying on paid cloud APIs.&lt;/p&gt;

</description>
      <category>devchallenge</category>
      <category>hf26challenge</category>
    </item>
    <item>
      <title>STAGE LADDER - A Private, Offline Speaking Coach Built for a Friend</title>
      <dc:creator>Aishwarya</dc:creator>
      <pubDate>Mon, 05 Oct 2026 01:51:30 +0000</pubDate>
      <link>https://dev.to/aishwarya_15/stage-ladder-a-private-offline-speaking-coach-built-for-a-friend-12gp</link>
      <guid>https://dev.to/aishwarya_15/stage-ladder-a-private-offline-speaking-coach-built-for-a-friend-12gp</guid>
      <description>&lt;p&gt;&lt;em&gt;This is a submission for the &lt;a href="https://dev.to/challenges/hacktoberfest-weekend-2026-10-01"&gt;Hacktoberfest Weekend Challenge: Build for a Friend&lt;/a&gt;&lt;/em&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  What I Built
&lt;/h2&gt;

&lt;h2&gt;
  
  
  Demo
&lt;/h2&gt;

&lt;h2&gt;
  
  
  Code
&lt;/h2&gt;

&lt;h2&gt;
  
  
  How I Built It
&lt;/h2&gt;

&lt;h2&gt;
  
  
  Why Does Open Innovation Matter?
&lt;/h2&gt;

&lt;h2&gt;
  
  
  Prize Categories
&lt;/h2&gt;

</description>
      <category>devchallenge</category>
      <category>weekendchallenge</category>
      <category>hf26challenge</category>
      <category>hacktoberfest</category>
    </item>
    <item>
      <title>STAGE LADDER - A Private, Offline Speaking Coach Built for a Friend</title>
      <dc:creator>Aishwarya</dc:creator>
      <pubDate>Sun, 04 Oct 2026 16:20:28 +0000</pubDate>
      <link>https://dev.to/aishwarya_15/stage-ladder-a-private-offline-speaking-coach-built-for-a-friend-6mp</link>
      <guid>https://dev.to/aishwarya_15/stage-ladder-a-private-offline-speaking-coach-built-for-a-friend-6mp</guid>
      <description>&lt;p&gt;&lt;em&gt;This is a submission for the &lt;a href="https://dev.to/challenges/hacktoberfest-weekend-2026-10-01"&gt;Hacktoberfest Weekend Challenge: Build for a Friend&lt;/a&gt;&lt;/em&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  What I Built
&lt;/h2&gt;

&lt;h2&gt;
  
  
  Demo
&lt;/h2&gt;

&lt;h2&gt;
  
  
  Code
&lt;/h2&gt;

&lt;h2&gt;
  
  
  How I Built It
&lt;/h2&gt;

&lt;h2&gt;
  
  
  Why Does Open Innovation Matter?
&lt;/h2&gt;

&lt;h2&gt;
  
  
  Prize Categories
&lt;/h2&gt;

</description>
      <category>devchallenge</category>
      <category>weekendchallenge</category>
      <category>hf26challenge</category>
    </item>
  </channel>
</rss>
