On: Show HN: Oqoqo – build evals and custom benchmarks for real-world tasks

[haritha1313]

users have complete control of what a good output looks like. we have seen that adding criteria that is specific to what the expected output is, including at trajectory level, helps a lot to avoid false positives