All
Articles 128,520Blog Posts 133,388Tech Tutorials 33,187Research Papers 24,721News 18,232
⚡ AI Lessons
Dev.to · Tatsuya Shimomoto
🧠 Large Language Models
⚡ AI Lesson
6d ago
LLM-as-Judge Shouldn't Aggregate Scores: Binary Checks as Evidence, One Holistic Verdict
A design pattern for LLM-as-judge: collect evidence with binary Yes/No checks, pick one named verdict holistically, and never aggregate into a score. With a cop
Dev.to · Tatsuya Shimomoto
1w ago
Building an Autonomous Agent on an M1 Mac, by Choice
A 16GB M1 Mac and 9B-class small models are not a budget compromise — they are a chosen constraint that exposes design flaws and trains the ability to separate
DeepCamp AI