Hotnot AI Dating Assistant

The description states that Hotnot AI Dating Assistant is a privacy-first ChatGPT assistant built as a proof-of-concept for the OpenAI 2026 hackathon. It integrates with an existing dating platform called "Hotnot" and uses the MCP protocol to enable limited interaction between ChatGPT and the backend. The system allows users to complete a guided interview via ChatGPT, which then creates a structured AI profile used for matchmaking. The assistant is designed to be non-autonomous, acting as a communication guide rather than making decisions on behalf of the user.

Key commercial signals:

• Not evidenced: No revenue, customers or traction data.
• Not evidenced: No pricing model or business model details.
• The project appears to be a prototype built for internal testing and not yet launched publicly.
• The system is designed with strong privacy controls and limited data exposure.
• The team size is one person (Radovan Zvolský), suggesting early-stage development.

The single most important open question: Is this a standalone product or an integration feature? If it's an integration, what is the strategic value of the existing Hotnot platform to the AI assistant?

The description states that Hotnot AI Dating Assistant is:

• A privacy-first ChatGPT assistant
• Built as a proof-of-concept for the OpenAI 2026 hackathon
• Integrated with an existing dating platform called "Hotnot"
• Uses the MCP protocol to connect ChatGPT with the backend
• Designed to conduct guided interviews with users about relationship preferences
• Creates structured AI profiles from interview responses
• Provides limited candidate recommendations based on compatibility scoring

Inferred: The assistant is not a standalone dating service but an add-on feature built on top of existing Hotnot infrastructure.

The description states that the project was created to:

• Move online dating away from superficial profile browsing toward more thoughtful ways of getting to know people
• Conduct natural but guided conversations instead of long questionnaires
• Help users understand their dating preferences and discover compatible matches
• Act as a safe intermediary for communication between matched individuals

The claim evolution appears to be:

1. Initial positioning: Privacy-focused AI assistant for dating
2. Core value proposition: Structured interview process that leads to better compatibility matching
3. Future vision: Communication assistant that helps users draft messages and communicate safely

Inferred: The project is positioned as a privacy-preserving alternative to traditional dating platforms, with an emphasis on AI-guided compatibility rather than visual or demographic matching.

The description states:

• The target user is someone looking for compatible dating matches
• Users must be 18+ and verified accounts
• The system is designed for people who want to get to know others more deeply before making decisions about relationships
• The assistant helps users describe their needs more clearly and highlights possible shared values

Not evidenced: No specific customer segments, personas or ICP defined beyond general dating seekers.

The description states:

• Hotnot already existed as a web-based dating and communication platform
• The AI assistant is built on top of the existing Hotnot backend
• Access is granted only to permitted roles and accounts that are active, verified, and meet minimum age requirements
• The current version is a private proof of concept intended for ChatGPT Developer Mode and approved internal testers

Not evidenced: No pricing model, revenue streams or business model details provided.

The description states:

• Built with Go (backend), React/TypeScript (frontend)
• Uses PostgreSQL, Valkey, SeaweedFS
• Authentication handled by Ory Hydra
• Uses OAuth authorization-code flow with PKCE S256
• Implements MCP protocol for ChatGPT integration
• Data encrypted using AES-256-GCM
• Interview answers stored in encrypted form
• AI analysis done through OpenAI Responses API with provider-side storage disabled
• Strict validation of outputs to prevent sensitive data leakage
• Separation of concerns between different services and permissions

Inferred: Strong technical architecture focused on security, privacy, and isolation of components.

The description states:

• Current version is a private proof of concept
• Intended for ChatGPT Developer Mode and approved internal testers
• Integration built on existing Hotnot backend
• Access controlled through feature flags and permissions
• No public launch or user base mentioned

Not evidenced: No traction, adoption, or maturity metrics.

The description states:

• The project is positioned as an alternative to traditional dating platforms that rely heavily on profile photos and basic demographic information
• Focuses on compatibility based on deeper personal values and preferences rather than surface-level attributes
• Uses AI to guide users through structured interviews instead of traditional questionnaires

Not evidenced: No competitive landscape or market positioning details.

The description indicates:

• The system is a proof-of-concept, not yet launched publicly
• Only one team member (Radovan Zvolský) involved in development
• Limited functionality - no messaging or matching features currently implemented
• Heavy emphasis on privacy and data minimization may limit utility
• Integration with existing platform suggests dependency on Hotnot's success

Inferred: High risk of failure due to lack of traction, limited scope, and dependency on another platform.

1. What is the strategic relationship between this AI assistant and the existing Hotnot platform?
2. How does the team plan to transition from a proof-of-concept to a scalable product?
3. What are the specific plans for monetization or revenue generation?
4. How will user data be handled beyond the current privacy controls?
5. What is the timeline for full feature implementation and public launch?
6. How does the system handle edge cases in conversation flow or user behavior?
7. What are the technical challenges expected in scaling this solution?