Knight Rise: On-device chess: Rust engine, board scanner, zero tracking

Hey Product Hunt! 👋 I’m the CTO at Olib AI, and today we are launching Knight Rise. After spending the last couple of years building privacy-focused tools like Owl Browser and the StealthOS workspace, we realized the casual apps we use every day are doing the exact opposite. Most modern chess apps are packed with telemetry, force you to create cloud accounts, and process your game analysis on slow remote servers. We wanted to build something entirely different: a zero-dependency architecture. Here is what we packed onto the device: Native Rust Engine: The engine evaluates your game offline. Our early docs mention a 2300-ELO engine, but we are currently upgrading to a 2750 ELO benchmark. It absolutely destroys me, but it includes a Coach Mode that explains why my moves are terrible in plain English. Computer-Vision Scanner: Use your camera to snap a physical board or use the system share extension to parse a screenshot from another app, and the local pipeline digitizes it instantly. No photo uploads. Local P2P Multiplayer: We used an open-source mesh networking library to let you play someone across the table over local Wi-Fi with zero server lag. The Business Angle (Why $0 backend costs matter): Because we shifted all the intelligence and vision pipelines to the edge, our monthly server compute cost is exactly $0. We built this to show that businesses can adopt powerful AI layers without relying on external cloud connections or compromising user ethics. Play to Earn Pro: The core app is free. There are Pro features, but you don't need a credit card—you can unlock free months of Pro just by completing in-game achievement sets and doing the daily puzzles. Would love to hear your feedback on the board scanner, the engine speed, or the edge architecture! I'll be hanging out in the comments all day to answer your questions.