[LRUG] [JOBS] machine learning & ruby at Dressipi

[Frederick Cheung]

TLDR: Interested in (but not necessarily experienced with) a job involving machine learning? Email me

We're looking for a ruby developer to join the Dressipi machine learning team in London (near Oxford Circus) as a permanent member of staff and work on our recommendations engine  (the thing that predicts what items we should show people given all the data that we have). We're not expecting prior knowledge of machine learning or recommender systems, although clearly an interest in these areas is helpful.

The elevator pitch is that we're trying to solve personalization in the clothing retail industry, both online and (in the near future) in store. We're helping both retailers improve their sales and consumers find the right items for them and our recommendations engine is a key part of this. The recommendations team is still pretty small, so we've got quite a backlog of ideas that we think have great potential but haven't been able to try out yet. We're hiring an extra person to help unblock that pipeline and improve the service we provide to both our partner retailers and their consumers. We're working with some big retailers (M&S, Very, Topshop etc.) so your work will impact millions of consumers and how they find and buy clothing items.

This job mainly involves prototyping new recommenders, evaluating them and then bringing them to production. Some of the ideas we're kicking around at the moment involve inferring which attributes of a garment a user might like given the interactions they've had and trying to define similarity between garments and using that for recommendations. We discuss these ideas and approaches as a team and once we've converged on something we'd like to try, you'd start implementing one of the approaches, working with other team members as appropriate. We also work closely with our stylists who have some pretty deep domain specific knowledge.

The tech stack for our web apps is Rails, Elasticsearch, Mongodb + Mysql, but the recommendations engine gets all its data from a separate postgres db and doesn't have a frontend as such. The bulk of the recommendations engine is ruby, with key hotspots written in java & rust. We're open in principle to people bringing in new languages for these hotspots - as long as we don't end up with every developer using a different language!

There's a bit more info at dressipi.com/jobs but basically if you're interested email me and we'll take it from there.

Thanks for reading!

Fred

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lrug.org/pipermail/chat-lrug.org/attachments/20160419/c5638d1c/attachment.html>