<?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: Mr Robot</title>
    <description>The latest articles on DEV Community by Mr Robot (@mrrobotji).</description>
    <link>https://dev.to/mrrobotji</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%2F4141536%2F9acc9d03-7b16-490f-a17f-3f4339a13052.jpg</url>
      <title>DEV Community: Mr Robot</title>
      <link>https://dev.to/mrrobotji</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/mrrobotji"/>
    <language>en</language>
    <item>
      <title>Stop Building Stateless AI: Enter the Walrus Session 8 Hackathon ($2,500 Prize Pool) to Build Chatbots That Remember</title>
      <dc:creator>Mr Robot</dc:creator>
      <pubDate>Thu, 24 Sep 2026 15:29:14 +0000</pubDate>
      <link>https://dev.to/mrrobotji/stop-building-stateless-ai-enter-the-walrus-session-8-hackathon-2500-prize-pool-to-build-d3d</link>
      <guid>https://dev.to/mrrobotji/stop-building-stateless-ai-enter-the-walrus-session-8-hackathon-2500-prize-pool-to-build-d3d</guid>
      <description>&lt;p&gt;The current standard for Conversational AI is entirely transactional. A user logs in, asks a question, gets an answer, and leaves. When they return the next day, the system treats them like a complete stranger. &lt;/p&gt;

&lt;p&gt;As developers, we spend countless hours optimizing database schemas and crafting seamless state management for our web applications, yet we allow our AI integrations to remain completely stateless. We force users into a frustrating loop of constantly re-explaining their preferences, their past issues, and their account details. &lt;/p&gt;

&lt;p&gt;The next major leap in UX isn't just about faster models or larger context windows—it is about moving from transactional AI to &lt;strong&gt;relational AI&lt;/strong&gt;. Agents need persistent, decentralized state. &lt;/p&gt;

&lt;p&gt;If you are ready to tackle this infrastructure challenge, the &lt;a href="https://www.deepsurge.xyz/hackathons/c0141a4a-21be-4009-bc63-7c168608c849" rel="noopener noreferrer"&gt;DeepSurge Hackathon Portal&lt;/a&gt; is currently hosting the &lt;strong&gt;Walrus Session 8: Chatbots That Remember&lt;/strong&gt; developer challenge.&lt;/p&gt;

&lt;h3&gt;
  
  
  The Hackathon: Move Beyond the Blank Slate
&lt;/h3&gt;

&lt;p&gt;Running from &lt;strong&gt;September 18 to October 9&lt;/strong&gt;, this session challenges builders to eliminate the "blank slate" problem in AI by integrating decentralized memory. &lt;/p&gt;

&lt;p&gt;By leveraging the &lt;a href="https://github.com/MystenLabs/MemWal" rel="noopener noreferrer"&gt;Walrus Memory GitHub repository&lt;/a&gt;, developers can build or retrofit a chatbot that actively stores and recalls context across multiple sessions, devices, and days. The goal is to prove that when an AI retains long-term user state, the entire interaction paradigm changes.&lt;/p&gt;

&lt;p&gt;You can apply this to almost any domain:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;  &lt;strong&gt;Customer Support:&lt;/strong&gt; A widget that remembers a user's open ticket from yesterday.&lt;/li&gt;
&lt;li&gt;  &lt;strong&gt;Internal Tools:&lt;/strong&gt; An HR or IT bot that knows an employee's department and past requests.&lt;/li&gt;
&lt;li&gt;  &lt;strong&gt;Gaming:&lt;/strong&gt; An NPC that holds grudges or remembers past alliances.&lt;/li&gt;
&lt;li&gt;  &lt;strong&gt;Education:&lt;/strong&gt; A tutor bot that tracks a student's struggles over an entire semester.&lt;/li&gt;
&lt;/ul&gt;

&lt;h3&gt;
  
  
  Submission Criteria for Builders
&lt;/h3&gt;

&lt;p&gt;To qualify for the prize pool, you need to deliver a working implementation with proven real-world utility:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Integration:&lt;/strong&gt; You must use Walrus Memory to persist and recall context. &lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Real-World Proof:&lt;/strong&gt; The system must be tested by at least &lt;strong&gt;3 distinct users&lt;/strong&gt;, generating a minimum of &lt;strong&gt;10 memories per user&lt;/strong&gt;.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Deployment:&lt;/strong&gt; Your chatbot must live somewhere accessible—a web widget, a Discord server, Telegram, Slack, or a CLI tool.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Open Source:&lt;/strong&gt; A public GitHub repository with clear setup instructions is required.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;The Builder's Log:&lt;/strong&gt; You must publish a 500–800 word technical write-up on Medium or Inkray detailing your build process. The judges want to see the exact moment your bot leveraged a past memory to improve the conversation, as well as any friction you experienced along the way.&lt;/li&gt;
&lt;/ol&gt;

&lt;h3&gt;
  
  
  The $2,500 Stablecoin Prize Pool
&lt;/h3&gt;

&lt;p&gt;Whether you are a seasoned AI engineer or just starting to tinker with local models, there are multiple tracks to win:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;  &lt;strong&gt;Best Overall Chatbot ($900 Total):&lt;/strong&gt; Awarded to the top three projects that demonstrate the cleanest code, best documentation, and most impactful use of persistent state (1st: $500, 2nd: $250, 3rd: $150).&lt;/li&gt;
&lt;li&gt;  &lt;strong&gt;Beyond the Big Two ($300 Total):&lt;/strong&gt; A dedicated track for developers who step away from standard Anthropic or OpenAI APIs. Build with open-weight models (Llama, DeepSeek, Mistral) via Ollama/vLLM, or use alternative providers like Google Gemini or Groq to claim one of two $150 prizes.&lt;/li&gt;
&lt;li&gt;  &lt;strong&gt;Best Technical Write-up ($300 Total):&lt;/strong&gt; Three $100 prizes for the most honest, clear, and helpful articles documenting the developer journey.&lt;/li&gt;
&lt;li&gt;  &lt;strong&gt;Bug Bounty ($500 Total):&lt;/strong&gt; Find reproducible bugs or friction points in the Walrus Memory SDK? Log them on GitHub for a chance at five $100 bounties.&lt;/li&gt;
&lt;li&gt;  &lt;strong&gt;Community Promotion ($500 Total):&lt;/strong&gt; Five $100 prizes for developers who help expand the reach of the hackathon by sharing it in external developer communities and forums.&lt;/li&gt;
&lt;/ul&gt;

&lt;h3&gt;
  
  
  How to Enter
&lt;/h3&gt;

&lt;p&gt;It is time to stop building bots that start from zero every single time. &lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Register Now:&lt;/strong&gt; Head to the &lt;a href="https://www.deepsurge.xyz/hackathons/c0141a4a-21be-4009-bc63-7c168608c849" rel="noopener noreferrer"&gt;DeepSurge Hackathon Portal&lt;/a&gt; to lock in your spot.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Review the Docs:&lt;/strong&gt; Dive into the &lt;a href="https://docs.wal.app/walrus-memory/examples/chatbot" rel="noopener noreferrer"&gt;Chatbot Reference Architecture&lt;/a&gt; to understand how Walrus Memory handles persistent state.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Join the Hacker Community:&lt;/strong&gt; Hop into the &lt;a href="https://discord.com/invite/walrusprotocol" rel="noopener noreferrer"&gt;Walrus Discord&lt;/a&gt; to brainstorm, troubleshoot, and team up.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;Submit your final project and article via the &lt;a href="https://airtable.com/appoDAKpC74UOqoDa/shro5iVzzjoWfZlPK" rel="noopener noreferrer"&gt;Official Airtable Form&lt;/a&gt; by &lt;strong&gt;October 9th&lt;/strong&gt;.&lt;/p&gt;

</description>
      <category>agents</category>
      <category>ai</category>
      <category>backend</category>
      <category>softwaredevelopment</category>
    </item>
  </channel>
</rss>
