Twitter has announced the results of a public contest to find bias in its photo cropping system.

The bounty challenge was launched in July after Twitter users showed that the site’s automated cropping tool favored the faces of people with lighter skin tones over those with darker skin tones, raising some questions about how the software prioritized skin color and certain factors over others.

The challenge was to figure out what other errors and biases the cropping system might have, and then fix the problems.

First place went to Bogdan Kulynych, whose entry showed how beauty filters can circumvent the algorithm’s scoring model, which in turn reinforces traditional beauty standards. The entry showed that the algorithm favored young and slim faces with light or warm skin tones. Kulynych won $3,500.