We run an AI startup, and are surrounded by a highly gender-biased environment with regards to developers. Men earn 181% of what women earn on average in the tech industry, and we think this is despicable, given we have female developers on our team.
What it does
We built hequals to solve this issue, by using AI to intelligently allocate tasks in a project or work environment by analysing team members skills. It uses Github Repository commit information to make an informed decision on who should be working on what task, with 0 bias.
How we built it
We used Python Flask to create the webserver, and used the Github Developer API to retrieve repo data. From this, we use a combination of NLTK, proprietary semantic processing algorithms, as well as Google Cloud to predict the most appropriate team member for each issue in a Git repo.
Challenges we ran into
It was a real struggle to build the entire algorithm and interface in the time of #codeathon, so we had to pull an all nighter in order to finish the algorithm.
Accomplishments that we're proud of
Being some of the younger participants but managing to complete a solution during #codeathon that we have actually been using ourselves is something that we're very proud of. It's a feeling that we have been encouraging our peers to aim for, and we're in the process of managing our own Hackathon in order to facilitate this.
What we learned
We learnt that sleep is fundamental to success. We could've finished the whole platform during the second morning, and the grogginess from the all-nighter was detrimental to our programming efficiency.
What's next for hequals (h=)
We're currently using hequals for task allocation in our main project Questo, and it has been working quite well. We're planning to continue using it internally, and take the project further by distributing it to other startups in Singapore, to propagate gender equality in the brutal developer environment. :)