We publish machine learning research on Bitcoin and the Lightning Network, then ship it. Every paper below powers a live system moving real value on the Amboss platform.
3
Published papers
$16M+
Liquidity allocated in production
4,640
Channel opens executed by our RL agent
3-6%
Probing efficiency gain in production
We address liquidity placement in the Bitcoin Lightning Network (LN): given a fixed budget, which channels should a node open to maximize its routing capacity?
Deployed in production for peer recommendations: 4,640 channel-open decisions allocating 267.3 BTC (over $16 million) across 30 managed nodes.
We present Bayesian Binary Search (BBS), a novel framework that bridges statistical learning theory/probabilistic machine learning and binary search.
In production for Lightning channel balance probing with a 3 to 6% efficiency gain. Published in the peer-reviewed journal Algorithms, in collaboration with Stillmark.
This research explores the feasibility of using machine learning models to interpolate channel balances within the network, which can be used for optimizing the network's pathfinding algorithms.
Predicts hidden channel balances from node and channel features alone, outperforming the equal-split baseline by 10%.
All of this research is running in production, helping us automate nodes and liquidity across the cluster. Every model we publish makes real decisions on live infrastructure, moving real value every day. Nothing here sits on a shelf.