Show HN: Lexe – Self-Custodial Lightning Nodes in TEEs, Python and Rust SDKs

Show HN: Lexe – Self-Custodial Lightning Nodes in TEEs, Python and Rust SDKs(blog.lexe.app)

5 points by phlip9 55 days ago | 3 comments

Lexe is next-generation Bitcoin infrastructure: self-custodial wallets and SDKs that are easy to use, always online, and hosted for free. Your keys are protected by secure enclaves, allowing your node to stay online 24/7 and receive payments reliably.

New in this release:

- First ever public release of Lexe Wallet on the App Store and Google Play

- Human Bitcoin Addresses (BIP 353): receive Bitcoin at ₿yourname@lexe.app

- Python and Rust SDKs, published to PyPI and crates.io

- Open source node, app, and SDKs published at https://github.com/lexe-app/lexe-public

fangx 55 days ago |

Lightning has always been broken in one way or another.

In the early days of Lightning, users ran nodes at home, which costed $300 in hardware, plus managing channels and liquidity. Way too expensive in terms of both time and money. Mobile wallets like Breez and Phoenix made things easier, but 'offline receive' was unreliable: payments would still fail if your phone was in power-saving mode, had bad reception, or was off the grid. Custodial wallets fixed that, but now you had to give up your keys.

@phlip9 and I got sick of all this and wanted a Lightning wallet that Just Works™ while retaining the self-sovereignty that Bitcoin is all about. After some iterating, we landed on the radical idea of running an entire Lightning node inside a hardware enclave (specifically, Intel SGX) in the cloud. The enclave isolates your node from the operators of the system - we can't read your keys. But because it's hosted, your node stays online 24/7.

This is our first ever public release, still in beta. We'd love to hear your feedback.

timpan 50 days ago | |

What’s the performance like on the SGX?

fangx 48 days ago | | |

Some things are extremely slow - getting the current time is literally 1000x slower than on regular Linux. But overall, since we're built on LDK, which is written in Rust, CPU isn't typically our bottleneck. Memory is a bigger problem because we have to store the entire Lightning Network graph in memory, but we amortize this cost across multiple user nodes with our "meganode" architecture.