AI quoting platform project- QC-326

Desired discipline(s): Engineering - computer / electrical, Engineering, Computer science, Mathematical Sciences
Company: Fülhaus
Project Length: 6 months to 1 year
Preferred start date: 09/01/2020
Language requirement: English with some French proficiency
Location(s): Montreal, QC, Canada; Canada
No. of positions: 1
Preferred institutions: Université de Montréal

Search across Mitacs’ international networks - check this box if you’d also like to receive profiles of researchers based outside of Canada: 

About the company: 

Fülhaus is the furnishing solution for the rental market. The Fülhaus Haus-in-a-Box introduces the most comprehensive interior design and furnishings packages that grow a rental business. We optimize the interior design to achieve higher ADRs for our clients, as our product line Haus-in-a-Box is everything they need to get their rental up and running with good design at the forefront. Haus-in-a-Box is a true white-glove service delivered end-to-end, as Fülhaus sources, furnishes, installs, stages, and professionally photographs every Haus-in-a-Box in as little as three weeks. Fülhaus also offers an innovative, risk-free rental option where clients receive premium design and furnishings for less up-front. With our rental package, clients receive a brand-new Haus-in-a-Box package every two years, with a free redesign and the option to convert to ownership at any time. Furthermore, everything in a Haus-in-a-Box is shoppable. Each package offers a tablet with access to the Fü online store featuring the same items in their space. When guests buy something from our online store during or after their stay, the client receives a commission.

Please describe the project.: 

The Fülhaus platform has been created to compile furniture and decor stock levels from our 300+ suppliers, and enable our designers to complete a furniture and decor package based on the size of the client's property and style selection. This currently requires an interior designer to sift through thousands of products in order to achieve the client's desired look, and all of the items in their package. Fulhaus launches 6 new Moodboards twice a year, and our design team then iterates on these moodboards everytime a package is ordered in that style. In order for us to produce these packages at scale, we are looking to develop a recommendation engine that will suggest a package to the Designer based on the moodboard selected. We would like to establish whether current machine learning technology is available and capable to produce a reasonable result  - i.e. a furniture package that resembles the moodboard. If this is possible, we would like to develop this recommendation engine further to eventually make it available to the public, whereby they may even be able to connect their pinterest or instagram feed, to produce a package tailored to them.

Required expertise/skills: 

What we’re looking for in a candidate:

  • Computer Science, Mathematics or equivalent
  • Very good proficiency in statistical platforms such as Python, R, SAS, or MatLab and any other basic libraries for machine learning such as scikit-learn and pandas
  • Proficiency in building AI models in deep learning frameworks such as Keras, TensorFlow, or Theano.
  • Experience on an artificial intelligence project is a plus
  • Expertise in visualizing and manipulating big datasets
  • Proficiency with OpenCV
  • Familiarity with Linux
  • Ability to select hardware to run an AI/ML model with the required latency


  • Skills in web technology are a plus (JavaScript)
  • Experience in dealing with data and different APIs
  • Experience with NoSQL databases (MongoDB)
  • Have experience with wireframing tools (e.g., Figma)
  • Familiarity with Git and GitHub