HackerTrans
TopNewTrendsCommentsPastAskShowJobs

landonxi

no profile record

Submissions

SWE-Bench Failures: When Coding Agents Spiral into 693 Lines of Hallucinations

surgehq.ai
22 points·by landonxi·10 ay önce·1 comments

Extracting text from a pdf broke ChatGPT

surgehq.ai
7 points·by landonxi·10 ay önce·2 comments