OpenRig - Peer to Peer donation based hardware sharing platform
This is a submission for Weekend Challenge: Generosity Edition What I Built I grew up teaching myself to code in Pakistan on hardware that struggled to run modern dev tools. GPU hours on Lambda Labs cost $0.50–$3/hr. For a student trying to fine-tune a model or run a training job, that's not accessible. OpenRig is a peer-to-peer compute sharing platform. Donors register their idle hardware and…
This submission is part of the Weekend Challenge: Generosity Edition, where the creator built OpenRig, a peer-to-peer compute sharing platform. In traditional GPU cloud services, costs can range from $0.50 to $3 per hour, making it inaccessible for students fine-tuning models or running training jobs. OpenRig enables donors to register their idle hardware, which runs a lightweight agent binary.
Recipients describe their needs in plain English, and the platform uses Google Gemini AI to match them with the most suitable hardware based on social and educational impact. Once approved, the donor spins up an isolated Docker container with SSH access, which is then connected through a bore tunnel. The recipient receives SSH credentials in their dashboard to connect to the resource.
The donor can revoke access at any time, automatically tearing down the container and tunnel. The platform utilizes a relay architecture to prevent exposing donors' IP addresses, with bore acting as an open-source TCP tunnel. Docker provides resource-capped containers, while Google Gemini 3.6 Flash scores the requests for impact, ensuring donors prioritize impactful use cases rather than making random selections.
The agent architecture is a standalone Go binary that polls the backend every 3 seconds to fetch provisioning and revocation data. The main challenge was ensuring donors' real IPs remain hidden, achieved through the relay architecture. The project was built in 48 hours using Go, Fiber v3, SQLite, SvelteKit, Tailwind CSS, Supabase, Google Gemini, and bore.
The creator notes plans for future improvements, such as replacing Docker with Firecracker for proper VM isolation, hosting the bore server, adding session expiry enforcement, proper bore process monitoring, and making the agent a proper system service.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
