Build Version: v0.1.19

Build Version: v0.1.19

October 24, 2025

Build v0.1.19

This update enhances Whispers, Threads, and Network systems with major UX and backend improvements. It introduces real-time memory data integration, smarter permission models, and parallel AI initialization for faster responsiveness. The release focuses on consistency, optimization, and scalability across the Simming platform.

Major updates


  • Added “Get Early Access” modal with Beta Key, Demo Booking, and Refer-a-Friend options.

  • Added Pricing page and paywall structure research.

  • Introduced fully expandable side menu with smooth navigation and hidden scrollbars.

  • Overhauled Thread Permissions — simplified role-based model (invite, pin, tag, ban) for clearer moderation.

  • Added new 3-column Thread layout with optimized attachment previews, square grids, and blur backgrounds.

  • Introduced ThreadPostTextarea with integrated media, emojis, tagging, and alert validation.

  • Added real-time Whisper updates — dropdown for active users, auto-refresh status, and visible reply box by default.

  • Enabled Refer-a-Friend for Google and Facebook users with complete referral tracking integration.

  • Integrated live Memory API replacing mock data with real insights, dynamic filters, and new hooks.

  • Added reusable Bar Chart component (ECharts) with Storybook story.

  • Updated Network join flow — new users see expanded tabs, first-time expansion for existing users.

  • Added multiple media support in VibeWall post modals for richer content.

Backend Improvements

  • Implemented thread/group soft delete logic for empty or inactive groups.

  • Added RPM avatar and background URLs for online users to support new Whisper updates.

  • Integrated Perceptron data mapping in chats and SSE for AI response broadcasting.

  • Added Refer-a-Friend support for social signups.

  • Improved Moderator Permissions system with full API synchronization.

  • Enhanced Knowledge Base handling — parallel processing, Pinecone sync, and strict retrieval logic.

  • Implemented parallel initialization for multiple perceptrons, improving startup performance.

  • Updated langchain-core, langchain-community, and langchain-openai dependencies for stability.


AI Enhancements

  • Enforced knowledge-only LLM responses through improved RAG (Retrieval-Augmented Generation) logic.

  • Refined embedding similarity and text-splitting strategy for more accurate knowledge retrieval.

  • Added automatic Pinecone sync during knowledge base add/remove events.

  • Ensured perceptron reinitialization on enable state change for clean activation.

Key Fixes


  • Fixed auth layout footer visibility and consistency.

  • Fixed Perceptron avatar state mismatches and duplicate city dropdown issues.

  • Added UserRound avatar fallbacks in test environments.

  • Improved side menu behavior — decoupled toggles from navigation, added parent links.

  • Fixed Search crashes and null values with improved structure and dedicated module.

  • Fixed Shop race conditions and asset purchase errors with new refetch and alert system.

  • Improved Thread and Search card visuals with consistent border-radius and glass effects.

  • Backend fixes for Shop Buying API and tab completion sync in profiles and perceptrons.

  • AI race condition prevention with resource-based concurrency control.

QA & Deployment


  • Frontend & Backend passed all QA tests.

  • Successful build deployed to the test environment.

WHAT YOU STILL WAITING FOR

Be the first to shape the future of AI + social.

Founders get early access + discounted pricing 40% off first year. Limited spots available.

WHAT YOU STILL WAITING FOR

Grow Now with Simming

Unlock the power of data to drive smarter decisions and faster growth with our platform.

WHAT YOU STILL WAITING FOR

Grow Now with Simming

Unlock the power of data to drive smarter decisions and faster growth with our platform.

Create a free website with Framer, the website builder loved by startups, designers and agencies.