Show HN: Open Wearables – self-hosted wearable data platform with an AI layer

Show HN: Open Wearables – self-hosted wearable data platform with an AI layer(github.com)

2 points by bartmichalak 119 days ago | 3 comments

Hi, I'm Bart, one of the maintainers of Open Wearables.

For the past few months we've been building an open-source, self-hosted backend that normalises health and fitness data from multiple wearable providers (Garmin, Whoop, Apple Health, Samsung Health and others) into a single AI-ready REST API - and thought it was worth sharing here too :)

We built this out of our own experience doing custom wearable integrations for clients, and from working with paid SaaS solutions that often lack the flexibility to customise data pipelines or debug what's actually happening under the hood. Surprisingly, there weren't any mature open-source projects filling this gap.

What you get:

- A single normalised API covering 60+ health metrics (cardiovascular, sleep, body composition, activity, respiratory) and 80+ workout types across all supported providers

- A dashboard to manage users, API keys, and sync status — with a full view of each user's health data across connected devices

- An AI layer via MCP, letting you plug users' health data into any LLM or build a custom agent around it

- Open-source mobile apps for SDK-based providers (like Apple Health) so you can quickly test how data flows from device to your platform end-to-end

On the roadmap: more providers, deeper AI features, direct wearable connections via BLE/ANT+ and performance and stability improvements to support enterprise deployments.

Happy to answer any questions. If this solves a problem you've been dealing with - give it a try, open an issue or contribute. All feedback welcome!

adola_ring_2026 113 days ago |

This is great. We're building Adola (https://adola.app), a subscription-free smart ring with a similar focus on data ownership. One of the biggest hurdles we've found is that even if the hardware is solid, the vendor-locked APIs (Oura/Whoop) make it really hard for users to actually 'own' their biometrics.

Our goal is to give users direct access to the raw PPG and sensor data without it being gated by a cloud subscription. Integrating with a platform like Open Wearables is exactly the kind of ecosystem we'd love to support.

bartmichalak 113 days ago | |

Hey! Your ring looks very promising, especially the subscription-free model and the focus on raw sensor access. Curious about the data access side on your end - are you planning to expose the PPG/sensor data via an SDK, a REST API, or both? I'd want to understand the data flow to figure out the best integration path (e.g. whether we'd pull from a local SDK on-device or sync from a cloud endpoint) with Open Wearables.

blockpilot_ai 113 days ago | |

Interesting project.

I've been thinking a lot about tools for organizing long AI conversations. Curious how people here currently manage them.