atypica.AI Changelog

Track real-time market trends and consumer signals. AI-powered pipeline collects and scores trending topics by relevance, then surfaces them in an interactive marketplace. Browse what's hot across categories, filtered by your locale.

Upgrade your subscription plan directly from your account with a live price preview showing the prorated difference. Plus, you can now choose between an affiliate discount or enter your own promotion code at checkout.

Studies, Panels, Personas, and Interview Projects can now be archived and restored. Keeps your workspace clean without permanently deleting anything.

Group multiple AI Personas into reusable research panels. Launch group discussions, one-on-one interviews, and other research types directly from a panel. All outputs (reports, podcasts, etc.) are organized in one place. Reuse the same panel across projects for more efficient and consistent research.

Studies, AI Personas, interview projects, panels, reports, and podcasts are all searchable now. Finding past research and content is much faster.

Upload PDFs, images, and other reference materials during research and interviews. The AI actively reads and references these files throughout the analysis, grounding research in your existing materials.

Introducing user-level personal memory (previously only available for teams). Through a short conversation, the AI learns your background, role, and interests, then automatically applies this context to all future research. The more you use it, the better it understands you — no need to repeat yourself.

Redesigned team memory builder with a smoother interview flow. After completing the interview, the AI automatically suggests research topics tailored to your team to help you start your next study right away.

Each Persona now comes with auto-generated role tags and demographic attributes, plus a self-introduction. Easier to understand who each Persona represents at a glance.

Research reports have been redesigned with a typography-first approach, emphasizing readability and visual hierarchy. Reports are now cleaner and more professional.

All list pages (studies, personas, interviews, panels) now share a consistent card-style layout with pagination for a smoother browsing experience.

Launched complete MCP Study API with 9 core research tools. Developers can integrate atypica's research capabilities into their applications through standard MCP protocol. Added atypica-research skill for using research features through MCP clients like Claude Desktop. Download skills from skill0.io. See MCP documentation for setup and usage.

Full AWS Marketplace integration allowing enterprise customers to subscribe directly through AWS accounts. Supports automatic registration, login, and subscription management to simplify enterprise procurement.

New database-driven research template system where AI recommends research scenarios based on your role. Supports dynamic generation and auto-refresh of templates to help you quickly start research.

Sage AI experts can now be exported as Claude Skills, allowing your custom experts to serve you in more scenarios. Sage is now officially featured in site navigation for easier discovery and access.

View AI's deep reasoning process during research to understand how AI thinks and reaches conclusions. Makes research more transparent and helps you understand AI's decision logic.

Memory system now supports delete and replace operations for more precise personal information management. Optimized memory triggers: automatically updates memory after research completion without manual intervention. Added quality gates to ensure high-quality and relevant memory content.

Research reports and podcasts can be shared to Twitter/X with one click. Optimized report page SEO and WeChat sharing compatibility for smoother sharing experience.

Added bilingual product documentation system to help users better understand product features. Admin dashboard now includes blog management interface supporting content creation and publishing. Enhanced About page showcasing complete product features and technical evolution.

Established persistent memory system where AI automatically remembers user preferences, habits, and contextual information for personalized service. Two-layer memory architecture (Core Memory + Working Memory) with intelligent reorganization and compression. New user capabilities page to view and edit personal memory anytime. Supports both user-level and team-level memory management. As conversations grow, AI understanding deepens without requiring repeated explanations.

Brand new research initiation experience: AI automatically understands your needs, determines the most suitable research type (Product Testing/Market Insights/Content Creation/Strategic Planning, etc.), and displays a complete research plan. Confirm with one click to start, no repeated back-and-forth needed. Added AI-generated research scenario shortcuts tailored to your role (Product Manager, Marketer, Startup Owner, Creator, Consultant, Influencer) to inspire your research direction. All research types now support advanced features like referencing existing research, uploading files, and MCP tool integration. Upgraded to Claude Sonnet 4.5 model for more accurate understanding and faster response.

Added design philosophy documentation: "The Less AI, The More AI" core principle. Optimized report and podcast generation prompts to emphasize research findings over methodology. Cover generation shifted from fixed style to dynamic selection. Instagram migrated to v2 API. Notification system migrated to Intercom event tracking.

Added discussionChat as alternative to interviewChat, supporting group discussion scenarios. discussionChat for viewpoint collision and solution testing (3-8 participants), interviewChat for deep personal insights (5-10 participants). AI automatically selects the most appropriate research method during planning phase based on objectives. Completed studyLog architecture migration from database fields to message-driven architecture, where all research content flows through messages for unified flexibility.

Brand new fast research mode focused on rapidly generating high-quality podcast content. Five-stage automated workflow: Topic Understanding → Podcast Planning → Deep Research → Podcast Generation → Completion. Dual-model collaboration using Claude 3.7 Sonnet and Gemini 2.5 Pro. Intelligent podcast content strategy and search strategy planning, automatically generates opinion-oriented podcast scripts and audio.

Added cinematic intro animation effects and progress bar to display research progress. Supports quick jump to report functionality, optimized visual hierarchy and console display.

Database performance optimization (foreign key indexes, N+1 query fixes, Token deadlock resolution). System enhancements (interview project import/export, mobile menu, security updates, etc.). Code refactoring and cleanup for improved system stability and maintainability.

Implemented Deep Research functionality as an MCP (Model Context Protocol) service. Supports multiple research experts: Grok Expert, Trend Explorer, and more. Integrated Perplexity Sonar Pro for high-quality web searches. Provides standard MCP API interface for third-party integration with streaming output support.

Users can create and manage personal API Keys. Teams support multi-member API Key management with permission validation and ownership control. Deep Research MCP endpoint supports API Key authentication.

Reports and podcasts support AI auto-generated cover images (integrated Gemini 2.5 Flash Image model). Supports custom aspect ratio and layout controls, podcast RSS supports platform-specific cover ratios.

New "Product R&D" research template focusing on market trends, user needs, and creative ideation. Optimized research workflow and report structure for product innovation scenarios. Helps teams quickly validate product ideas and market opportunities.

Launched Super (personal) and SuperTeam (team) unlimited token plans. Unlimited token usage for high-intensity research needs. Flexible team member management and subscription controls.

Comprehensive Pricing page overhaul with enterprise feature showcase. Enterprise-exclusive features: custom development, dedicated customer success, API access, etc. Optimized enterprise lead collection with enhanced information gathering.

Platform achieved SOC 2 security compliance certification. Certification badges displayed across website, pricing, and enterprise pages. Enhanced enterprise-grade security and compliance assurance.

Officially launched friends.atypica.ai affiliate platform. Integrated Tolt tracking system for referral links and commission management. Automatic tracking of referral sources for user signups and payments.

Launched evolvable AI expert agent system. Two-layer memory architecture (Core Memory + Working Memory) supporting long-term knowledge accumulation. Intelligent knowledge gap identification with automatic supplementary interviews. Builds continuously evolving domain experts through continuous learning and knowledge supplementation. Supports version control and knowledge source management for expert knowledge traceability.

Research reports can now be converted into podcast audio with one click, supporting online listening and downloading. Multiple podcast styles available: deep dive analysis, opinion discussion, and debate format. Intelligent script generation with automatic host dialogue configuration. Email notifications sent when podcast generation is complete. Full functionality available to Pro, Max, and Team plan subscribers.

Added multiple-choice question support for richer interview formats. Interview reports can be generated, shared, and downloaded as PDF. Share interview projects via invite links with configurable expiration. Set preferred form of address for interviewees. Multi-language interview support (Chinese and English). File attachment support in interview conversations.

Personas can now be shared via public access links. Persona chat supports file uploads as conversation context. Improved search and filtering with tag and preview features.

New AI-powered recommendations for research next steps. Continue deeper exploration based on existing research with reference context. Unified management interface for reports and podcasts for easier access and sharing.

AI engine upgraded for improved response speed and stability. Enhanced multi-model support optimizing cost and performance balance. System architecture optimizations for better overall user experience.

Previously, AI research jumped straight into interviews and search. Now every study begins with a visible plan that outlines who will be contacted, what will be investigated, and why each step matters.

The planning step automatically selects the right business analysis frameworks (JTBD, STP, GE matrix, etc.) so teams understand the strategic lens behind the upcoming interviews and desk research.

Before the final report is generated, the AI prepares a detailed research log that summarizes collected information and key decisions to give context behind the conclusions.

Both features now available to all plans (removed preview/beta status)

• •Free/Pro plans: Access to limited features
• •Max/Team plans: Access to full functionality

• •AI automatically analyzes your interview project brief and optimizes questions for better research outcomes
• •Transforms complex questions into clear, answerable segments following research best practices
• •Applies fact-first methodology and generates granular questions to capture deeper insights
• •Works seamlessly with both human interviews and AI persona interviews

• •Enhanced voice input speed and responsiveness
• •Redesigned persona import workflow with better status visibility

Added team workspaces, supporting member invitations, removal, and role management. Implemented user identity switching for seamless work between personal and team spaces. Introduced dedicated routes and navigation menus for the team edition.

Launched the core AI Persona Import feature, opening up a public beta for Max plan users. Users can upload their own interview transcripts (PDF, JSON, CSV) to build completely private Human AI Personas (Proprietary). The AI provides multi-dimensional analysis and scoring for the uploaded content, generating an interactive AI Persona and visualizing the analysis with tools like radar charts.

Integrated Groq Whisper API for near real-time, streaming transcription, dramatically improving the fluency of voice interviews (2025-07-04).

Completed the core development of the new Interview Project and moved it into the internal Alpha testing phase. The new workflow aims to simplify the creation and sharing process for interviews (2025-07-24).

Added a dedicated attachments table to manage user-uploaded files centrally, enabling file reuse and optimizing storage (2025-07-20).

Completed the core database migration from China to the US region, laying the foundation for global services (2025-06-29). Launched the new new-study-interview feature, clarifying user research goals through an initial AI conversation. Published the first global launch announcement on Twitter, officially marking the product's entry into the international market.

Completed the core development of the Product R&D research flow and moved it into the internal Alpha testing phase (2025-06-27).

Launched a completely redesigned official website (atypica.ai) with an improved information architecture and visual experience, featuring all AI-generated illustrations (2025-06-27).

Launched the MAX subscription plan, offering more tokens and advanced features (2025-06-12). Implemented plan upgrades/downgrades with automated pro-rated pricing based on remaining tokens (2025-06-26). Transitioned to USD-based pricing and integrated Stripe's Alipay auto-debit feature to streamline payments (2025-06-19).

Integrated Tavily for real-time web searches, providing richer context during the initial research phase. Added Twitter as a new data source for building more globally relevant user personas. Introduced the "Real Persona" concept, allowing personas based on actual interviews to be highlighted during research.

Implemented automatic classification of research queries (Testing, Planning, Insights, Creation) to match them with distinct research flows and report structures (2025-06-09).

Integrated Google Imagen 4.0 and Midjourney to dynamically generate high-quality, context-aware images directly within research reports (2025-05-25).

Enabled prompt caching for Bedrock Claude. By hashing historical messages, it reduces token consumption for repetitive requests by 90% (2025-05-29).

Optimized the voice recognition feature using the Groq Whisper model for instant speech-to-text conversion, significantly improving interaction speed. Fixed a bug where prompt caching failed due to an issue in the Vercel AI SDK.

Users can now upload documents (PDF, etc.) as context when initiating a study. The AI references this information during subsequent interviews and report generation (2025-05-18).

Integrated the pgvector extension into the database, adding Embedding vectors to Personas to enable semantic, similarity-based searches (2025-05-13).

Deployed a standalone html-to-pdf service, allowing research reports to be downloaded as high-quality PDF files with a single click (2025-05-12).

The Study Agent can now pass more specific instructions and context to the Interview Agent, receiving richer feedback and significantly reducing overall token consumption.

Successfully migrated the backend database from MySQL to PostgreSQL, paving the way for advanced features like vector search and complex queries (2025-04-29). Resolved initial performance issues post-migration by adding indexes to nerd stats queries.

Introduced the gpt-4o model with support for parallel tool calling to effectively reduce research costs (2025-04-16). Adopted Google Gemini 1.5 Flash for specific tasks to further optimize costs (2025-04-21). Implemented strict token and step limits to prevent research tasks from running over budget or time (2025-04-28).

Switched the billing system from simple "points" to a more granular Tokens-based model. Officially launched the Pro subscription plan and a new Pricing page (2025-04-20). Integrated Stripe to support global credit card payments (2025-04-09).

Split ChatMessage data into a separate table, resolving data override issues during concurrent writes and improving system stability and scalability (2025-04-12).

Added TikTok and Instagram as new data sources to broaden consumer insights (2025-04-17).

Introduced a user points system and launched the first payment feature ("buy a coffee") via the Heidian API. Added a Hello Agent to capture enterprise leads through conversation.

Added an invitation code system for early user onboarding (2025-03-31).

Overhauled the multi-step process into a unified "research-driven chat" interface, where users can kick off a full study just by asking a question. Introduced the Study Agent as a "commander" to plan tasks, orchestrate specialized agents, and report back to the user.

Adopted a split-screen design (left: chat, right: atypica llm console) to provide real-time transparency into the AI's thought and execution process (2025-03-19). Implemented a "Replay" feature with shareable links (2025-03-22). Added "Nerd Stats" for tracking token and step consumption (2025-03-26). Added a light/dark theme toggle (2025-03-25). Generated the first changelog page automatically from git log.

Launched a multi-page, guided scout → personas → analyst → interview → report process, establishing a foundational framework for market research.

• •Analyst: Create and manage research topics, stored in the database.
• •Interview: Simulate AI-driven interviews with Personas.
• •Report: Automatically generate HTML reports with a shareable link.

Introduced user authentication (login/signup) and made research history private (2025-03-16). First deployment to an AWS cluster at atypica.musedam.cc, with GitHub CI for continuous integration.

Product concept defined: build interactive User Personas from social media data to test product innovation ideas (2025-03-05).

Achieved the first end-to-end flow: automatically summarizing and creating User Personas from "Xiaohongshu" data. The SavePersona tool successfully saved the first AI-generated persona to the database.

• •Next.js + Vercel AI SDK + LLM Tools as the core framework.
• •Claude 3.7 Sonnet as the primary reasoning model.
• •Significantly reduced early token costs by rewriting prompts with an experimental method (2025-03-08).