ChoiceCapture by ChoiceOS

Note: This analysis is based solely on the self-reported project description provided by the caller. All claims are unverified and should be treated as stated by the author only.

The description states that ChoiceCapture is:

• An AI-assisted live-commerce capture and reconciliation system
• Designed to observe a live-selling event and create dependable records of every product run, including sold items, unsold items, and Buy It Now purchases
• A modular architecture with components such as:
  • Chromium browser extension for event observation
  • Local capture and event-processing service
  • Product-run and sale state machine
  • Structured event and evidence storage
  • GPT-5.6 product matching
  • Live event dashboard
  • OBS-compatible browser-source overlay
  • CSV and evidence exports
  • Post-live reconciliation workflow

It separates platform listing titles from actual product identities, using AI to match generic listings like “Item on Screen 147” with real inventory items such as “CHO-FUN-0147”.

Claim: ChoiceCapture is a system that captures live-commerce events and reconciles them with inventory records.

Evidence: Author's own write-up.

The author states:

• The product addresses an operational problem in live-commerce where sellers use generic listings like “Item on Screen 147” instead of real product names.
• Existing workflows rely on manual methods (spreadsheets, screenshots) that are slow and error-prone.
• ChoiceCapture aims to automate this process using AI-assisted matching and structured data capture.

It positions itself as part of a broader "ChoiceOS commerce operating system", suggesting it may be intended for integration into larger platforms or ecosystems.

Claim: ChoiceCapture solves the problem of disconnected transactional information and real product identity in live-commerce.

Evidence: Author’s own write-up.

The description states:

• The primary use case is for live-commerce sellers, particularly those running high-volume events like auctions
• It was inspired by a need within Choice Collectibles Ltd
• It supports platforms where generic listings are common (e.g., eBay-style auction sites)
• It targets users who want to generate accurate sales records, inventory reconciliation data, and fulfillment-ready files

No explicit segmentation beyond this is described.

Claim: The target customer is live-commerce sellers running high-volume events with generic product listings.

Evidence: Author's own write-up.

Not evidenced. No mention of pricing models, monetization strategies, or business model assumptions in the description.

Finding: No evidence of business model or pricing structure.

The system uses:

• .NET 9, ASP.NET Core, React, TypeScript, JavaScript, HTML5, CSS3
• Browser extension (Chromium-based)
• Docker containerization
• GPT-5.6 for product matching
• Computer vision and OCR as fallbacks
• REST API, WebSockets, JSON, SQLite
• OBS browser-source overlay
• GitHub Actions for CI/CD

Key technical features include:

• Event-driven state machine to track auction progress
• Structured event and evidence storage
• Prevention of false sales via stability checks
• Support for varying auction lengths (1 sec to 30 mins)
• Handling of Buy It Now purchases separately from auctions
• Preservation of original evidence and timestamps

Claim: ChoiceCapture uses a modular, event-driven architecture with AI-assisted matching.

Evidence: Author's own write-up.

Not evidenced. The project is described as a Build Week prototype submitted to a hackathon (Devpost). There is no mention of:

• Customers
• Revenue
• Adoption
• Product-market fit
• Prior versions or iterations

The next steps include piloting within Choice Collectibles Ltd, but no external traction is reported.

Finding: No evidence of traction or maturity beyond prototype stage.

Not evidenced. The description does not reference competitors or similar tools in the live-commerce capture space.

Finding: No competitive context provided.

• Unproven commercial viability: No evidence of external adoption, revenue, or customer feedback.
• AI dependency risk: Reliance on GPT-5.6 for product matching introduces uncertainty about accuracy and scalability without real-world validation.
• Limited scope: The system is described as a prototype with no indication of platform integrations beyond internal use.
• Single-founder team: Only one member listed (Darren Mutch-Blache), which may limit execution capacity.
• Hackathon origin: Submitted to a hackathon, suggesting early-stage development and untested assumptions.

Inference: The lack of traction or commercial validation raises concerns about product-market fit and scalability.

1. What is the current stage of development beyond the Build Week prototype?
2. Has Choice Collectibles Ltd piloted ChoiceCapture? If so, what were the results?
3. Are there any external users or partners currently testing the system?
4. How does the AI matching perform in practice — what are typical confidence scores and error rates?
5. What platforms is it designed to integrate with beyond its current scope?
6. Is there a plan for monetization or commercial deployment?
7. What are the key assumptions about user behavior and platform compatibility that need validation?

Note: These questions aim to uncover whether the self-reported claims reflect real-world usage or just internal experimentation.