We're Sponsoring the OOP Healthcare Hardware Hackathon

Momentum is a tool sponsor at Out of Pocket's Healthcare Hardware Hackathon, April 17-19 in San Francisco. This post explains why we decided to sponsor and what teams building there will have access to through Open Wearables openwearables.io.

What the OOP Hackathon Is

Out of Pocket runs one of the more credible healthtech communities in the US. Their content is specific, their audience is technical, and they don't confuse healthcare innovation with pitch decks. The hardware hackathon is a focused event: around 15 teams of 5 people, 36-48 hours to build health applications using wearable hardware and AI tools. Demo Day is April 19, open to the public.

Teams are expected to build something that works. The wearable hardware is provided at the venue. The software infrastructure is up to the teams, which is where we come in as a tool sponsor.

What Hackers Get from Open Wearables

Open Wearables is an open-source health intelligence platform. It handles wearable data collection, normalization, and health scoring so developers can build features instead of rebuilding the data layer. The platform currently supports 9 providers and over 300 devices, with a unified data model that returns the same API response structure regardless of which device a user wears.

For the hackathon, each team gets a dedicated hosted instance, pre-configured and pre-loaded with sample health data. No local Docker setup, no provider OAuth to configure from scratch. Teams arrive with an API key and a portal login. The quickstart guide targets 15 minutes from zero to first API call.

Beyond raw data, the platform includes open algorithms for health scoring: sleep quality, recovery, HRV, strain, and VO2 max. The algorithms are readable and forkable. A team building a clinical triage tool can audit exactly how a recovery score is computed, adjust thresholds for their population, and produce results that are explainable to clinicians.

The platform also ships an MCP server, which gives any LLM structured access to health data and reasoning. A team building an AI health coach can connect Claude or GPT-4 to a user's wearable history, health scores, and trend analysis without writing prompt engineering from scratch to handle raw sensor data.

After the hackathon, top projects that built with Open Wearables will be featured on openwearables.io, with a two-hour technical consultation from the OW team to help take the project further.

What Teams Can Build in 36 Hours

The resource card we prepared for event teams describes four categories:

AI health coach. Pull sleep, recovery, and strain data from multiple wearables in a single API call, feed it to an LLM via the MCP server, and return personalized recommendations grounded in real scores. No surveys, no self-reporting.

Recovery monitoring tool. Aggregate data from multiple devices into a single health score, add alerting when metrics drop outside a user's baseline range, deploy self-hosted. The multi-device normalization is already handled by the platform.

Clinical triage or scheduling tool. Factor a patient's wearable health scores alongside their EHR data so care teams make decisions with the full picture. The health scoring algorithms are auditable, which matters in clinical contexts where the reasoning behind a recommendation has to be traceable.

Personal health assistant. Users ask in plain language and get answers grounded in real wearable data rather than generic advice. The MCP layer handles the translation between natural language queries and structured health data.

None of these projects require teams to build the data collection layer from scratch, which takes 6-12 months of engineering effort when done provider by provider.

Why We're Sponsoring This

Developers who build a working application with a platform over a 36-hour period understand what it can and cannot do better than any demo we could give them. If a team at OOP builds something real with Open Wearables and it holds up under time pressure, they carry that experience back to their company. If it breaks, we learn what needs fixing.

Open Wearables exists because we kept rebuilding the wearable data layer for health app clients and decided the better approach was to open-source it. Six companies are currently using it in production. The hackathon is the most direct external test we have run: put the platform in front of developers with no prior context, give them a deadline, and see what gets built.

Deploying Open Wearables takes a few minutes. What takes longer is convincing a development team it's worth adding to their stack. A working hackathon demo is a better argument than a sales deck.

Bartosz Will Be On-Site

Bartosz Michalak, our Director of Engineering, will be on-site both Saturday and Sunday as a technical mentor. He built significant parts of the OW codebase and knows where the rough edges are.

His job at the hackathon is to help teams scope their projects, unblock them when the API behaves unexpectedly, and redirect when they try to build something too complex for 36 hours. The goal is working demos. Our team will also be available via the event support channel during hacking hours for remote questions.

Attending Demo Day

Demo Day is April 19 and open to the public. OOP will be announcing registrations shortly. If you're in San Francisco and want to see what developers build when given real health infrastructure and a hard deadline, it is worth attending.

If you want to try Open Wearables ahead of the event, the quickstart guide is on GitHub (github.com/the-momentum/open-wearables).

The FAQ on themomentum.ai covers provider support, self-hosting requirements, and what "open algorithms" means in practice.