From 714da5e98a17b79f7ead881beb02e95f263f4828 Mon Sep 17 00:00:00 2001 From: cah Date: Wed, 25 Feb 2026 20:33:54 -0700 Subject: [PATCH] docs: add Apache 2.0 license and AI disclosure Co-Authored-By: Claude Opus 4.6 --- LICENSE | 2 +- docs/ai-disclosure.md | 28 ++++++++++++++++++++++++++++ 2 files changed, 29 insertions(+), 1 deletion(-) create mode 100644 docs/ai-disclosure.md diff --git a/LICENSE b/LICENSE index 261eeb9..03a4a90 100644 --- a/LICENSE +++ b/LICENSE @@ -186,7 +186,7 @@ same "printed page" as the copyright notice for easier identification within third-party archives. - Copyright [yyyy] [name of copyright owner] + Copyright 2026 Corey Hahn Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. diff --git a/docs/ai-disclosure.md b/docs/ai-disclosure.md new file mode 100644 index 0000000..359f845 --- /dev/null +++ b/docs/ai-disclosure.md @@ -0,0 +1,28 @@ +# AI Disclosure + +This project was developed with assistance from Claude (Anthropic), an AI coding assistant. + +## How AI Was Used + +- **RTL design and debugging**: AI assisted with SystemVerilog implementation of the LDPC decoder core, Wishbone interface, and Caravel integration wrapper. AI-driven verification found and fixed 7 RTL bugs through standalone and vector-driven testbenches. +- **Behavioral modeling**: Python simulation model (density evolution, BER analysis, test vector generation) co-developed with AI assistance. +- **Build system and tooling**: OpenLane configuration, SDC timing constraints, and Makefile setup guided by AI. +- **Documentation**: Architecture docs, project reports, and contest submission materials drafted with AI assistance. +- **Design space exploration**: Code rate comparison, base matrix optimization, SC-LDPC analysis performed with AI-guided Python scripts. + +## Human Contributions + +All architectural decisions, algorithm selection (layered offset min-sum, QC-LDPC rate 1/8, IRA staircase structure), target application (photon-starved optical communication), and final design review were made by the human designer. The AI served as a coding assistant, technical reference, and verification tool. + +## Transparency + +Full conversation transcripts with the AI assistant are available upon request. Key AI-assisted files include: +- `verilog/rtl/*.sv` — RTL implementation +- `../model/*.py` — Python behavioral model (in parent ldpc_optical repo) +- `openlane/*/config.json` — Synthesis configuration +- `docs/` — Documentation + +## Tool + +- **AI Model**: Claude (Anthropic) via Claude Code CLI +- **Usage Period**: February 2026