{"slug": "walrus-sessions-8-building-chatbots-that-actually-remember-you-2500-in-prizes-9", "title": "Walrus Sessions 8: Building chatbots that actually remember you ($2,500 in prizes, ends Oct 9)", "summary": "A developer spent several days auditing MemWal, the open-source SDK behind Walrus Memory, reading through its auth layer, on-chain contract, and sample apps, and reported a couple of real bugs that maintainers fixed quickly. The work is tied to Walrus Sessions 8, a live hackathon session on building or retrofitting chatbots that persist context across sessions, users, and devices, with $2,500 in prizes ending October 9. The developer highlighted the SDK's constant-time auth checks, on-chain delegate-key verification, and quarantine/replay protection as a way to understand secure persistent agent memory under the hood.", "body_md": "Most chatbots forget everything the second a conversation ends. Ask it \n\nsomething on Monday, come back Tuesday, and you're a stranger again.\n\n**Walrus Sessions 8: \"Chatbots That Remember\"** is a live hackathon \n\nsession tackling exactly that — building (or retrofitting) a chatbot \n\nthat uses **Walrus Memory** to persist context across sessions, users, \n\nand devices.\n\nBuild or retrofit a chatbot that:\n\nAny use case counts: customer support, onboarding, tutoring, sales, \n\ncommunity bots, even game NPCs.\n\nI spent a few days going deep on the underlying open-source SDK \n\n(MemWal) — not just building with it, but actually reading through the \n\ncodebase: the auth layer, the on-chain contract, the sample apps. It's \n\na genuinely well-built system (constant-time auth checks, on-chain \n\ndelegate-key verification, careful quarantine/replay protection), and \n\ndigging into *how* persistent memory is implemented securely taught me \n\nmore than most tutorials do. Ended up finding and reporting a couple of \n\nreal bugs along the way, which the maintainers fixed fast.\n\nIf you've ever wanted an excuse to actually understand how persistent, \n\nencrypted agent memory works under the hood instead of just calling an \n\nAPI — this is a good one to dig into, whether you build the full \n\nchatbot or just go bug-hunting in the SDK.", "url": "https://wpnews.pro/news/walrus-sessions-8-building-chatbots-that-actually-remember-you-2500-in-prizes-9", "canonical_source": "https://dev.to/walgo/walrus-sessions-8-building-chatbots-that-actually-remember-you-2500-in-prizes-ends-oct-9-3i16", "published_at": "2026-09-24 02:19:07+00:00", "updated_at": "2026-09-24 02:59:31.099976+00:00", "lang": "en", "topics": ["ai-agents", "ai-tools", "developer-tools", "ai-infrastructure"], "entities": ["Walrus Memory", "MemWal", "Walrus Sessions 8"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/walrus-sessions-8-building-chatbots-that-actually-remember-you-2500-in-prizes-9", "markdown": "https://wpnews.pro/news/walrus-sessions-8-building-chatbots-that-actually-remember-you-2500-in-prizes-9.md", "text": "https://wpnews.pro/news/walrus-sessions-8-building-chatbots-that-actually-remember-you-2500-in-prizes-9.txt", "jsonld": "https://wpnews.pro/news/walrus-sessions-8-building-chatbots-that-actually-remember-you-2500-in-prizes-9.jsonld"}}