Tag
This article outlines the pitfalls of using OpenRouter for accessing open-source AI models, based on the author's experience with 18 million messages, including significant benchmark variations across providers and issues with vision models.
A tweet discusses how AI agents slow down over time because their vector databases are optimized for static data, but real agents continuously add new data, causing performance degradation.
AI agents often fail in production because they lack access to real business data, internal docs, and live customer context. To succeed, they need actual content, live data integration, clear handoffs, and human oversight.
A discussion of potential issues when connecting Obsidian notes to AI, including privacy risks and data loss.
This paper presents a systematic analysis of evaluation pitfalls in multimedia event extraction, identifying issues such as inconsistent data processing, inconsistent task assumptions, and overly relaxed evaluation settings that can lead to overestimated performance.