sidethink

Company: sidethink

Self-reported basis: The analysis is based entirely on the author's own description of the project, submitted as part of a Devpost entry for the OpenAI 2026 hackathon. No external verification or independent sources are available.

What it appears to be: A macOS application that listens to real-time voice or text during meetings and generates structured insights — such as missing discussion points, fact-checking claims, and creative thinking support — using AI-powered natural language processing and reasoning.

What changed: The project is described as a beta version built with Codex (a code generation tool) and deployed via FastAPI backend. It was submitted to a hackathon, suggesting early-stage development.

Single most important open question: Is there sufficient evidence of user need or adoption to justify further investment or development beyond the author's personal use case?

The description states that sidethink is an app that:

• Receives real-time voice or text during meetings.
• Uses speech-to-text (via Apple’s local STT or Soniox API).
• Converts the transcribed text into “claims” — clear, organized opinions.
• Generates “insight cards” from these claims to find missing discussion points, fact-check claims, or support creative thinking.
• Sorts insight cards by importance for quick reference during live meetings.

The app is built as a macOS application and has two modes:

• BYOK (Bring Your Own Keys), allowing users to control their data via API keys.
• A backend server built with FastAPI for on-demand use and enhanced reasoning.

Inference: The product appears to be a real-time meeting assistant that leverages AI to enhance discussion quality, but it is not yet proven in the market or adopted by users beyond the developer’s own use.

The author states:

• The inspiration came from wanting to identify missing points or verify claims during meetings.
• The app aims to help guide discussions more effectively and support better decision-making.
• It is described as a "realtime thinking buddy" for meaningful discussion.

Inference: The positioning is that of a productivity tool aimed at individuals in meeting-heavy roles, with an emphasis on real-time AI assistance. However, no evidence exists of how this compares to existing tools or whether it addresses a widespread pain point.

The description does not clearly define the target customer or ideal customer profile (ICP). The author mentions:

• Using a laptop for online meetings.
• Building for macOS users.
• Noting that the app is currently in beta and tested internally.

Inference: The likely user base is individuals who attend frequent remote meetings, particularly those using macOS. However, no evidence of actual users or market segmentation exists.

The description does not mention:

• A pricing model.
• Revenue streams.
• Monetization strategy.
• Customer acquisition plans.

Inference: No business model is evident from the self-reported description. The app appears to be a personal project, not yet monetized or intended for sale.

The author states:

• Built using Codex 5.6 Sol Extra High.
• Backend server built with FastAPI.
• Frontend built in Swift (macOS).
• Uses Apple’s local STT or Soniox API for transcription.
• The app has two modes: BYOK and backend use.

Inference: The technical stack suggests a lightweight, developer-focused tool. However, no evidence of scalability, performance, or robustness is provided.

The description states:

• The project is in beta.
• It was submitted to the OpenAI 2026 hackathon.
• The author built it using Codex and refined workflows.
• A TestFlight build is planned for internal closed beta testing.
• Golden test datasets were created, but more data is needed.

Inference: There is no evidence of user adoption or revenue. The product is in early development and has not yet been tested with real users beyond the developer’s own use.

The description does not mention:

• Competitors.
• Market analysis.
• How sidethink differs from existing tools (e.g., meeting summarizers, AI assistants).

Inference: No competitive positioning or market differentiation is evident. The author does not reference existing solutions in the space.

• No traction or revenue: The project is described as a beta version with no evidence of user adoption.
• Unproven value proposition: The app’s utility depends on generating “meaningful insights,” but there is no data to support this.
• Developer-centric tool: Built for personal use, not yet designed for broader market appeal or scalability.
• Limited technical depth: Reliance on Codex and GPT-based tools may limit control over output quality or long-term maintainability.
• No monetization strategy: No indication of how the product will be sold or funded.

1. What specific user problems are you solving, and how do you know they exist?
2. Have you conducted any user research or interviews to validate demand?
3. How do you plan to scale beyond a single developer’s use case?
4. What is your roadmap for monetization or product-market fit?
5. What are the key limitations of the current AI tools (e.g., Codex) that affect this product?
6. How will you collect and improve data for better reasoning over time?