Based on a recent report there are 59M freelancers in US. The freelance workforce is global and very competitive; the talent supply is higher than the jobs demand. You can read more about this changing global marketplace and the covid impact in this financial results presentation.

Freelancers face multiple challenges two of them being:

• research the demand for their skills/services
• difficulty in finding and winning projects

Freelancer.com has an excellent API that can be used to extract project details. It is a rich data set with a respectable history (more than 5 years); it contains both contests and projects, although I will only cover the latter.

The above mentioned reports contain a lot of good information (freelancer profiles, average bids per project, gross market volume trends, etc.). In this post I will try to answer the following question: can a project probability to be awarded be determined accurately?

Given the extensive schema I have been focusing on 2 main components:

• project details
• users details

#freelance #freelancing #tensorflow #deep-learning #freelancers