[Feature Request] Interactive In-App AI Market Research & Structured User Feedback Mechanism

Feature Overview & Community Request

Community member @Mordred proposed implementing an interactive, AI-driven conversational market research survey on chat.z.ai (similar to Anthropic's in-app interview pop-up model).

Key Motivations & Problem Statement

1. User Segmentation: Current user groups (Coders vs. Creative Writers vs. Roleplay/Companionship users) have distinctly different requirements for context retention, system prompts, content safety heuristics, and model behavior.
2. Structured Feedback Collection: Rather than relying on scattered thread messages, an interactive in-app conversational survey allows Z.ai to gather structured, high-signal qualitative data directly from active users.
3. Addressing Model Perception & Shifts: Post-GLM-4.6 releases have seen varying feedback from non-coding segments (e.g. Roleplay / Creative Writing). A guided feedback pipeline helps backend engineers consolidate patterns across large user volumes rather than reacting to isolated individual voices.

Proposed Implementation

• In-App Interactive Pop-up on chat.z.ai: An optional, conversational interview session conducted by an agent when users log into the platform.
• Segment-Specific Question Routing: Dynamically adapts questions based on whether the user primarily uses GLM for coding, creative writing, or API tool integration.
• Structured Backend Aggregation: Automatically categorizes responses into feature requests, tone/behavior feedback, and UX friction points for the engineering team.

Regards,
Roman M.

• Contributor / Ambassador at Z.ai
• The Vibe Coder's Blog on Telegram:
  https://t.me/VibeCodePrompterSystem

The proposal targets the chat.z.ai product surface and describes an optional conversational survey, segment-specific question routing, and backend response aggregation, but names no repository files or tests. Start by checking whether this feedback repository contains implementation code or only product proposals. Done would require an agreed scope and a defined path for delivering and validating the in-app survey.