Keepword

The description states that Keepword is:

• A Telegram bot.
• An AI assistant for team commitments.
• Designed to watch group chats and identify actionable promises.
• A system that requires confirmation before turning a promise into a follow-through action.
• Not a full task manager, but a tool to ensure conversation-based commitments are not lost.

Inferred: It uses a combination of Telegram Bot API (via Grammy), LLMs (OpenRouter), PostgreSQL (via Drizzle ORM), and Fastify for backend logic. A static Next.js landing page is also part of the stack.

The description states:

• Keepword aims to bridge the gap between “we agreed on this in the chat” and “someone actually follows through.”
• It is described as a lightweight memory for team conversations.
• The product does not replace existing tools but ensures that promises made in chat do not get lost.

Inferred: The positioning has evolved from a general idea of improving follow-through to a specific, privacy-conscious tool that integrates into Telegram and avoids noise by requiring confirmation.

The description states:

• Keepword is for teams using group chats (specifically Telegram).
• It targets users who make promises or assignments in chat but do not follow up.
• Chat administrators can configure settings like mode, language, timezone, and digest time per group.

Inferred: The target customer is small to mid-sized teams that use Telegram for collaboration and want to improve accountability without adopting a new tool.

Not evidenced.

The description does not state anything about pricing, monetization, or business model.

The description states:

• Built with TypeScript, Fastify, Telegram Bot API (Grammy), PostgreSQL via Drizzle ORM, and OpenRouter.
• Uses a lightweight pre-filter before sending messages to the LLM.
• Backend modules include Telegram updates, commitment extraction, confirmation, reminders, digests, onboarding, and privacy operations.
• A background worker runs reminder and digest jobs every minute.
• Private onboarding, reminders, overdue alerts, and digests are implemented.
• Supports English, Russian, and Spanish.

Inferred: The system is designed with a modular backend, idempotent delivery, and server-authorized callbacks to protect group boundaries. It handles timezones and language preferences as part of its core functionality.

Not evidenced.

The description does not mention any users, customers, revenue, or usage metrics beyond the hackathon submission. The project is described as a prototype built for a hackathon.

Not evidenced.

The description does not compare Keepword to other tools or identify competitors in the market.

• No traction or user feedback: The product is described as a hackathon submission with no evidence of real-world usage.
• Unproven commercial viability: No pricing, monetization, or business model is described.
• Limited scope for growth: The focus on Telegram and chat-based commitments may limit scalability or adoption outside that ecosystem.
• Founder team size: Only one member (Amirzhan Zhampeissov) is listed, which may indicate limited execution capacity.

1. What real-world feedback have you received from teams using Keepword beyond the hackathon?
2. How do you plan to scale beyond a single founder and a prototype?
3. Have you identified any specific use cases or industries where this tool might be more valuable?
4. What is your strategy for monetization or product-market fit validation?
5. How do you plan to handle privacy compliance (e.g., GDPR, data retention) at scale?