Side Job: Looking for someone to make predictions (NLP, Tabular, image) and combining them on real retail data

Hi guys,

Let me first introduce myself. My name is Jasper and I am the founder of a company that collects market data in Europe in order to help retail organizations better promote and price their products. In my spare time I like to test a bit with new technology and that’s how I started with the online Fast.ai course 1.5 years ago.

My company has collected market data (products, prices, images, anything else we could find :-)) for approximately 6 years now and have created a large database with different kinds of data. This data has been structured as much as we could and later on labeled by humans to categorize it and make it comparable over different retailers.

While being able to make simple classifiers based on Fast.ai to predict the right category for a product; we are now at the challenge where we would like to combine different kinds of data (tabular, NLP and maybe image too) to make this prediction much more accurate. However, as it is getting more technical and more time consuming at this point (I spend most of my time going to clients and directing our development team) I figured that it might be nice to post an open invitation for a Fast.AI student to test a bit with our data and see if we can jointly get this prediction to a higher level.

In exchange of making use of our (real world) data and a financial compensation for the work done, I would really love to get into contact with a student to see if we can get this going. If you are interested; I am more than happy to share some of the work I did; but also some of our more strategic directions. Work could be abroad (or if you are based in the Netherlands in real life too :-)).

Looking forward to hear from you,
Jasper

P.S. If this is posted in the wrong section or if it’s not appropriate, my apologies. I couldn’t find a section where I could post this message so I thought this would be the best place. If not I am more than happy to move it :-).

1 Like