andrewgazelka··on Show HN: CivBench a long-horizon AI benchmark for multi-agent gameshey first of all cool product. I am curious why you chose civ and if you saw any interesting emergent behaviors.
andrewgazelka··on Show HN: File-based cache for slow Python functionsThis looks cool :). A while ago, I wrote something similar that analyzes bytecode and invalidates the cache if the bytecode changes.https://github.com/andrewgazelka/smart-cache
andrewgazelka··on BlazingMQ: High-performance open source message queuing systemIs there something like BMQ but in Rust? Seems interesting you mentioned it.
andrewgazelka··on Classifying customer messages with LLMs vs traditional MLMy understanding was training on ChatGPT output was against OpenAI ToS. Is this incorrect for this use case (training BERT)?