<?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: Raam Bhandarkar</title>
    <description>The latest articles on DEV Community by Raam Bhandarkar (@raam_8h).</description>
    <link>https://dev.to/raam_8h</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%2F4159054%2F305dca11-ff32-4082-9c2f-ce4842b370fa.jpg</url>
      <title>DEV Community: Raam Bhandarkar</title>
      <link>https://dev.to/raam_8h</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/raam_8h"/>
    <language>en</language>
    <item>
      <title>Challenge 3</title>
      <dc:creator>Raam Bhandarkar</dc:creator>
      <pubDate>Sun, 18 Oct 2026 04:56:41 +0000</pubDate>
      <link>https://dev.to/raam_8h/challenge-3-3fke</link>
      <guid>https://dev.to/raam_8h/challenge-3-3fke</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;h3&gt;
  
  
  Industrial AI Assistant — Your Factory's Knowledge at Your Fingertips
&lt;/h3&gt;

&lt;p&gt;Industrial teams often waste valuable time searching through lengthy SOPs, equipment manuals, and technical documents to find answers to simple but critical questions.&lt;/p&gt;

&lt;p&gt;I am building a &lt;strong&gt;Multimodal GenAI Assistant&lt;/strong&gt; that makes industrial knowledge accessible through a simple conversational interface.&lt;/p&gt;

&lt;p&gt;Users can ask questions in natural language and get clear answers grounded in approved industrial documentation, along with source references to help them verify the information. The long-term goal is to support scanned documents, diagrams, and equipment images as well.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Who would pay for it?&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Manufacturing companies and factories&lt;/li&gt;
&lt;li&gt;Maintenance and engineering teams&lt;/li&gt;
&lt;li&gt;Industrial plants and equipment operators&lt;/li&gt;
&lt;li&gt;Organizations managing large technical documentation libraries&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The business opportunity is straightforward: reduce the time spent searching for information, make documentation easier to use, and help teams access relevant knowledge when they need it.&lt;/p&gt;

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

&lt;p&gt;Demo coming soon.&lt;/p&gt;

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

&lt;p&gt;GitHub repository: Coming soon.&lt;/p&gt;

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

&lt;p&gt;The project is designed around &lt;strong&gt;Retrieval-Augmented Generation (RAG)&lt;/strong&gt;, combining document retrieval with generative AI.&lt;/p&gt;

&lt;p&gt;The planned workflow is:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Upload industrial SOPs, equipment manuals, and technical documents.&lt;/li&gt;
&lt;li&gt;Extract text from PDFs and scanned pages using document parsing and OCR.&lt;/li&gt;
&lt;li&gt;Divide the content into searchable sections and create embeddings.&lt;/li&gt;
&lt;li&gt;Retrieve relevant sections when a user asks a question.&lt;/li&gt;
&lt;li&gt;Use a language model to generate an understandable answer grounded in the retrieved information.&lt;/li&gt;
&lt;li&gt;Display source references so users can verify the answer against the original documentation.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;&lt;strong&gt;Planned technology stack:&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Frontend:&lt;/strong&gt; React&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Backend:&lt;/strong&gt; Python and FastAPI&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;AI architecture:&lt;/strong&gt; RAG and embeddings&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Knowledge retrieval:&lt;/strong&gt; Vector search with PostgreSQL/pgvector&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Document processing:&lt;/strong&gt; PDF parsing and OCR&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;AI models:&lt;/strong&gt; An open-weight language model evaluated for accuracy, hardware requirements, and deployment cost&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The focus is not just on generating answers. It is on making those answers traceable to the documentation they come from and acknowledging when the available evidence is insufficient.&lt;/p&gt;

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

&lt;p&gt;Industrial AI should not be limited to organizations that can afford expensive proprietary systems.&lt;/p&gt;

&lt;p&gt;Open innovation makes it possible to experiment with open-weight models, customize retrieval pipelines, inspect system behavior, and deploy components in environments with different privacy, cost, and infrastructure requirements.&lt;/p&gt;

&lt;p&gt;For this project, open-source tools can help reduce vendor lock-in, make the architecture more adaptable, and provide a path toward private deployment for organizations that cannot send confidential manuals to external services.&lt;/p&gt;

&lt;p&gt;It also creates an opportunity for developers to contribute improvements in document parsing, multilingual retrieval, evaluation, and safety mechanisms.&lt;/p&gt;

&lt;p&gt;My goal is to build something practical that can evolve through community feedback rather than remain a closed demonstration.&lt;/p&gt;

&lt;h2&gt;
  
  
  My Agent Session
&lt;/h2&gt;

&lt;p&gt;Not available yet.&lt;/p&gt;

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

</description>
      <category>devchallenge</category>
      <category>hf26challenge</category>
    </item>
    <item>
      <title>Challenge 2</title>
      <dc:creator>Raam Bhandarkar</dc:creator>
      <pubDate>Sun, 11 Oct 2026 09:48:45 +0000</pubDate>
      <link>https://dev.to/raam_8h/challenge-2-2e56</link>
      <guid>https://dev.to/raam_8h/challenge-2-2e56</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;p&gt;I built &lt;strong&gt;TouchGrass AI&lt;/strong&gt;, an AI-powered project designed to help people spend less time scrolling on their screens and more time exploring the real world.&lt;/p&gt;

&lt;p&gt;The idea is simple: instead of just telling people to reduce their screen time, the app gives them personalized outdoor challenges, such as taking a short walk, visiting a nearby park, exploring a new place, or spending time with friends without their phones.&lt;/p&gt;

&lt;p&gt;The AI helps generate activities based on a user's available time, interests, and preferences, making it easier to turn small daily actions into healthier habits.&lt;/p&gt;

&lt;p&gt;It's designed for students, developers, remote workers, and anyone who spends too much time in front of a screen.&lt;/p&gt;

&lt;p&gt;The goal isn't to replace technology. It's to use technology to help people reconnect with the world outside it.&lt;/p&gt;

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

&lt;p&gt;💻 **GitHub Repository:&lt;a href="https://github.com/Codimow/touch-grasas" rel="noopener noreferrer"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;The project is built with an open-source-first approach, keeping the code accessible so other developers can explore it, contribute improvements, and build on the idea.&lt;/p&gt;

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

&lt;p&gt;I approached this project with one question: &lt;strong&gt;Can AI help people spend less time using AI and technology?&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;The application combines an interactive interface with an AI-powered activity recommendation system. Users can provide their interests and available time, and the AI suggests practical activities they can complete offline.&lt;/p&gt;

&lt;p&gt;The planned architecture includes:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Frontend:&lt;/strong&gt; A simple, user-friendly interface for discovering activities and tracking completed challenges.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;AI layer:&lt;/strong&gt; An open-source language model to generate personalized outdoor challenges and suggestions.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Backend:&lt;/strong&gt; Handles user preferences, challenge requests, and application logic.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Habit-building features:&lt;/strong&gt; Encourages users to complete real-world activities rather than endlessly consume digital content.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;&lt;strong&gt;AI model and framework:&lt;/strong&gt; [Add the actual open-weight model, inference tool, and framework you used.]&lt;/p&gt;

&lt;p&gt;I focused on making the experience lightweight and practical. The AI should help users take action, not give them another reason to stay on their screens.&lt;/p&gt;

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

&lt;p&gt;Open innovation matters because tools designed to improve our lives should be accessible to everyone, not just people who can afford expensive APIs.&lt;/p&gt;

&lt;p&gt;Using open-source AI can give developers more control over their applications, more flexibility in choosing models, and opportunities to experiment with different approaches.&lt;/p&gt;

&lt;p&gt;It also makes the project easier for the community to inspect, improve, and adapt for different users and environments.&lt;/p&gt;

&lt;p&gt;For a project like TouchGrass AI, openness creates opportunities to experiment with local inference, reduce reliance on external services, and make AI-powered habit-building tools more accessible.&lt;/p&gt;

&lt;p&gt;Most importantly, the community can help improve the idea and make it useful for more people.&lt;/p&gt;

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




&lt;p&gt;&lt;strong&gt;Build with AI. Step outside. Touch grass. 🌱&lt;/strong&gt;&lt;/p&gt;

</description>
      <category>devchallenge</category>
      <category>hf26challenge</category>
    </item>
    <item>
      <title>QueueFlow</title>
      <dc:creator>Raam Bhandarkar</dc:creator>
      <pubDate>Sun, 04 Oct 2026 13:23:07 +0000</pubDate>
      <link>https://dev.to/raam_8h/queueflow-4ip3</link>
      <guid>https://dev.to/raam_8h/queueflow-4ip3</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;p&gt;I built an &lt;strong&gt;IoT-enabled Smart Queue Management System&lt;/strong&gt; for my friend &lt;strong&gt;Shiva&lt;/strong&gt;, who often has to deal with long and unpredictable queues at service centers.&lt;/p&gt;

&lt;p&gt;The system connects a physical IoT setup with a digital queue platform. When a person is detected by an &lt;strong&gt;IR sensor&lt;/strong&gt;, the ESP32 communicates with the user application and the phone announces:&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;&lt;strong&gt;“Please press the button for token.”&lt;/strong&gt;&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;When the physical button is pressed, a digital token is generated instantly. The user receives a virtual ticket showing their &lt;strong&gt;token number, current serving number, people ahead, and estimated waiting time&lt;/strong&gt;.&lt;/p&gt;

&lt;p&gt;The system also provides an admin dashboard for monitoring queue length, footfall, service counters, waiting times, and resource utilization in real time.&lt;/p&gt;

&lt;h3&gt;
  
  
  The core flow
&lt;/h3&gt;

&lt;p&gt;&lt;strong&gt;Person Detected → IR Sensor → ESP32 → Voice Prompt → Physical Button → Digital Token → Virtual Queue → Live ETA → Notification&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;The goal is simple: &lt;strong&gt;reduce unnecessary physical waiting and make queues predictable.&lt;/strong&gt;&lt;/p&gt;

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

&lt;p&gt;The demo showcases a small-scale working service-center model with:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;ESP32&lt;/li&gt;
&lt;li&gt;IR sensor for visitor detection&lt;/li&gt;
&lt;li&gt;Physical push button for token generation&lt;/li&gt;
&lt;li&gt;Phone-based user interface&lt;/li&gt;
&lt;li&gt;Virtual digital ticket&lt;/li&gt;
&lt;li&gt;Real-time queue position&lt;/li&gt;
&lt;li&gt;Estimated waiting time&lt;/li&gt;
&lt;li&gt;Admin dashboard&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;&lt;strong&gt;Demo Video:&lt;/strong&gt;&lt;br&gt;
  &lt;iframe src="https://www.youtube.com/embed/94vYVQu6Fhw" width="710" height="399"&gt;
  &lt;/iframe&gt;
&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Live Demo:&lt;/strong&gt;&lt;br&gt;
&lt;code&gt;[live.shivasoni.me](url)&lt;/code&gt;&lt;/p&gt;
&lt;h2&gt;
  
  
  Code
&lt;/h2&gt;

&lt;p&gt;The project is built as a connected IoT + web application.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;GitHub Repository:&lt;/strong&gt;&lt;br&gt;
&lt;/p&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/RA4m15" rel="noopener noreferrer"&gt;
        RA4m15
      &lt;/a&gt; / &lt;a href="https://github.com/RA4m15/Queue-Flow" rel="noopener noreferrer"&gt;
        Queue-Flow
      &lt;/a&gt;
    &lt;/h2&gt;
    &lt;h3&gt;
      queue flow is the solution for crowd management using both the software and hardware
    &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;QueueFlow — Real-Time Smart Queue Management System&lt;/h1&gt;
&lt;/div&gt;
&lt;p&gt;QueueFlow is an enterprise-grade, real-time Smart Queue Management System designed to eliminate physical waiting lines, balance service counter workloads, track physical footfall via IoT sensors, and provide live operational analytics.&lt;/p&gt;

&lt;div class="markdown-heading"&gt;
&lt;h2 class="heading-element"&gt;Architecture Overview&lt;/h2&gt;
&lt;/div&gt;
&lt;p&gt;QueueFlow is built as a unified distributed system sharing one central backend and database:&lt;/p&gt;
&lt;div class="snippet-clipboard-content notranslate position-relative overflow-auto"&gt;&lt;pre class="notranslate"&gt;&lt;code&gt;                  ┌─────────────────────────────────────┐
                  │    Flutter Customer Mobile App      │
                  │   (Token Booking, Status, Alerts)   │
                  └──────────────────┬──────────────────┘
                                     │ REST / WebSocket
┌──────────────────────────────┐     │     ┌──────────────────────────────┐
│       React Admin Panel      │◄────┼────►│   ESP32 / RFID / IoT Sensors  │
│   (Dashboard, Counters,      │     │     │   (Footfall Entry/Exit Count)│
│    Display Board, Analytics) │     │     └──────────────┬───────────────┘
└──────────────┬───────────────┘     │                    │ IoT Events
               │ REST / WebSocket    │                    │ (x-iot-secret)
               ▼                     ▼                    ▼
     ┌──────────────────────────────────────────────────────────┐
     │       Node.js + Express + Socket.IO Unified Backend       │
     │      (Authentication, Queue Engine, Event Broker)        │
     └────────────────────────────┬─────────────────────────────┘
                                  │
                                  ▼
                     ┌─────────────────────────┐
                     │   MongoDB Atlas Cloud   │
                     │  (Multi-Tenant Data)    │
                     └─────────────────────────┘
&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;
&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Backend Engine&lt;/strong&gt;…&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/RA4m15/Queue-Flow" rel="noopener noreferrer"&gt;View on GitHub&lt;/a&gt;&lt;/div&gt;
&lt;/div&gt;


&lt;p&gt;The repository contains the ESP32 firmware, queue-management backend, user interface, admin dashboard, and simulation/demo components.&lt;/p&gt;

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

&lt;p&gt;The system combines &lt;strong&gt;IoT hardware, a real-time queue engine, and a web/mobile interface&lt;/strong&gt;.&lt;/p&gt;

&lt;h3&gt;
  
  
  Hardware
&lt;/h3&gt;

&lt;ul&gt;
&lt;li&gt;ESP32 as the main IoT controller&lt;/li&gt;
&lt;li&gt;IR sensor for detecting arriving visitors&lt;/li&gt;
&lt;li&gt;Physical push button for token requests&lt;/li&gt;
&lt;li&gt;Wi-Fi communication between the ESP32 and application&lt;/li&gt;
&lt;/ul&gt;

&lt;h3&gt;
  
  
  Software
&lt;/h3&gt;

&lt;ul&gt;
&lt;li&gt;React.js / Next.js for the user and admin interfaces&lt;/li&gt;
&lt;li&gt;Backend API for queue and token management&lt;/li&gt;
&lt;li&gt;Real-time communication for live queue updates&lt;/li&gt;
&lt;li&gt;Text-to-speech for the voice instruction&lt;/li&gt;
&lt;li&gt;PostgreSQL for queue and service data&lt;/li&gt;
&lt;li&gt;REST/WebSocket-based communication between system components&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The MVP uses an explainable queue-estimation approach based on &lt;strong&gt;people ahead, average service time, and active counters&lt;/strong&gt;.&lt;/p&gt;

&lt;p&gt;The architecture is designed so that historical queue data can later be used for machine-learning-based waiting-time prediction and resource optimization.&lt;/p&gt;

&lt;h3&gt;
  
  
  Open-source AI
&lt;/h3&gt;

&lt;p&gt;For the intelligent part of the system, the architecture is designed around &lt;strong&gt;open-source AI/ML components rather than depending on a proprietary AI API&lt;/strong&gt;. The prediction and recommendation layer can be extended with open-weight models and local inference while keeping the core queue system lightweight and explainable.&lt;/p&gt;

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

&lt;p&gt;Open innovation makes it possible to build this project as a &lt;strong&gt;low-cost, adaptable system&lt;/strong&gt; rather than depending on expensive proprietary infrastructure.&lt;/p&gt;

&lt;p&gt;The ESP32, open-source software libraries, open communication protocols, and community-built tools allow the system to be modified for different environments such as:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Hospitals&lt;/li&gt;
&lt;li&gt;Banks&lt;/li&gt;
&lt;li&gt;Government offices&lt;/li&gt;
&lt;li&gt;Railway counters&lt;/li&gt;
&lt;li&gt;Universities&lt;/li&gt;
&lt;li&gt;Customer service centers&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;More importantly, the system can be deployed locally and customized according to the requirements of each service center.&lt;/p&gt;

&lt;p&gt;A queue-management system should not be locked into one vendor or one ecosystem. &lt;strong&gt;Open innovation makes the solution easier to experiment with, extend, and adapt to real-world public-service environments.&lt;/strong&gt;&lt;/p&gt;

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

&lt;ul&gt;
&lt;li&gt;&lt;strong&gt;Hacktoberfest Weekend Challenge: Build for a Friend&lt;/strong&gt;&lt;/li&gt;
&lt;li&gt;IoT / Hardware&lt;/li&gt;
&lt;li&gt;Open Source&lt;/li&gt;
&lt;li&gt;AI / Intelligent Systems&lt;/li&gt;
&lt;li&gt;Smart City / Civic Technology&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Team
&lt;/h2&gt;

&lt;p&gt;&lt;strong&gt;Team SHARPRAD-X&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Devraja&lt;/li&gt;
&lt;li&gt;Shiva&lt;/li&gt;
&lt;li&gt;Arpana&lt;/li&gt;
&lt;li&gt;Raam&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;This project was built as a practical experiment in connecting &lt;strong&gt;physical IoT interactions with a real-time digital service experience&lt;/strong&gt;.&lt;/p&gt;

&lt;p&gt;Instead of making people wait in a queue and repeatedly check their position, we wanted to answer a simple question:&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;&lt;strong&gt;What if the queue could come to you instead of you waiting in the queue?&lt;/strong&gt;&lt;/p&gt;
&lt;/blockquote&gt;

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