BounceBoard

BounceBoard is a self-reported local-first training workflow for trampoline gymnastics, built by a solo developer (Vincent Wack) using Apple devices and AI tools like Codex and GPT-5.6. The product enables athletes and coaches to record synchronized video, sensor data, and pose analysis from an iPhone, Apple Watch, and Mac, with the goal of creating shared evidence for training without replacing human coaching.

The author states that BounceBoard is a prototype built during a hackathon, using Apple Vision for local pose detection and integrating motion data from the Watch. It includes features like clock alignment, multi-pass recovery of poses, and human-reviewed labels to maintain trust in automated outputs.

The description states that BounceBoard is:

• A local-first trampoline training workflow across Apple Watch, iPhone, and Mac.
• It uses:
  • Apple Watch to record workout, accelerometer, and device-motion data.
  • iPhone to control practice, record synchronized video, align clocks, and receive Watch recordings.
  • macOS Review Hub to bring together video, sensor signals, pose analysis, and annotations into a timeline.
• The system supports:
  • Synchronized recording across devices.
  • Local Apple Vision-based 2D pose extraction.
  • Human review of skill intervals and athlete identity.
  • Automated suggestions that are not treated as final until confirmed by a coach.
• Data remains on user devices unless explicitly exported.

Inference: The system is designed to support a coach’s workflow, not replace it. It emphasizes shared evidence and documentation over automation.

The author states:

• BounceBoard is built by a coach, with Codex, for athletes and coaches.
• It aims to make training easier to document, with an automated training diary and trustworthy practice history.
• The system is not meant to replace a coach, but to give better shared evidence.

Inference: The positioning is that of a coaching assistant tool, not a standalone automation or performance analytics platform. It is framed as a way to improve documentation and communication, not to drive performance gains directly.

The description states:

• BounceBoard is built for athletes and coaches in trampoline gymnastics.
• The author has experience as both an athlete (national level) and coach.
• It is designed for practical use with everyday Apple devices (iPhone, Apple Watch, Mac).

Inference: The ICP appears to be coaches and athletes in niche sports, particularly those using Apple ecosystems. The product is not positioned for mass market or general fitness.

The description does not state:

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

Not evidenced: No evidence of a business model, pricing, or monetization strategy beyond the author’s own development use case.

The description states:

• Built with Codex and GPT-5.6 as development-time collaborators (not used at runtime).
• Uses Apple technologies:
  • HealthKit, Core Motion, WatchConnectivity, AVFoundation, Apple Vision, GRDB/SQLite, SwiftUI, Swift.
• Implements a two-pass recovery pipeline for pose detection using alternative orientations.
• Includes clock alignment, reliable transfer protocols, and human-reviewed labels.
• The system is designed to preserve reliability over performance.

Inference: The technical approach emphasizes reliability, privacy, and multimodal sensor fusion, with a focus on robustness in edge computing environments. It is not a cloud-based or AI-as-a-service solution.

The description states:

• BounceBoard was built during a hackathon (OpenAI 2026).
• The author has no external users beyond personal testing.
• No revenue, customers, or adoption data are provided.
• It is described as a working prototype, not a commercial product.

Not evidenced: No evidence of traction, user base, or commercial adoption. The project is self-reported as experimental and not yet deployed in real-world settings.

The description does not state:

• Any competitors.
• Market size or competitive landscape.
• Existing tools for trampoline training or gymnastics analytics.

Not evidenced: No competitive analysis or market positioning beyond the author’s own claims.

The description states:

• The system is not yet tested with real athletes and coaches, only in development.
• It is a solo-developer project (1-person team).
• Reliability was a major challenge, with issues around device synchronization and data loss.
• No large public datasets exist for trampoline gymnastics, so automation is limited.

Inference: Key risks include:

• Lack of real-world testing.
• Limited scalability or commercial viability due to niche market.
• Dependency on Apple ecosystem and developer tools (Codex, GPT).
• Risk of over-engineering for a small user base.

1. What is the expected adoption rate among coaches and athletes in trampoline gymnastics?
2. How does BounceBoard plan to scale beyond a solo developer’s prototype?
3. Are there any plans to expand beyond Apple devices or integrate with other platforms?
4. What are the key assumptions about user behavior that underpin the product design?
5. Has the author considered how to monetize this tool, if at all?
6. How does BounceBoard handle data privacy and consent in real-world usage?