Dayjob: Tax – evaluating AI agents on 50 professional tax assignments s surgehq.ai 263 landonxi 3 days ago 174 comments
SWE-Bench Failures: When Coding Agents Spiral into 693 Lines of Hallucinations s surgehq.ai 22 landonxi 1 year ago 1 comments
Research trends in privacy, security and cryptography m microsoft.com 1 landonxi 3 years ago 0 comments
