Samuel Dillon

Principal ML Engineer at MATCHESFASHION

Sheffield, England, United Kingdom

About

I am an ambitious and motivated mathematics graduate, with a range of experience with different technologies and excellent analytic, communication and literacy skills, who is keen on finding good solutions to hard problems. I have had a couple of years of industry experience with Java and Scala, principally working on learning algorithms and optimization systems on both Windows- and Unix-based systems, feeding in data acquired from databases using SQL and HiveQL, and analysing results using R and spreadsheets (Excel and OpenOffice). I have also had university experience with C++, HTML, MATLAB/Octave and Maple, as well as working on my own small coding projects in all of these languages. I am also familiar with a wide variety of mathematical areas, such as number theory, group theory, graph theory, automata theory, differential geometry, topology and functional analysis.

Experience

  • Principal ML Engineer at MATCHESFASHION
    Aug 2020 - Present · 5 yrs 11 mos

    Working in the recommendations team, I have built recommendation engines for multiple purposes. I am also in charge of ML Ops as well as sharing my knowledge with other/junior members of the team.

  • Machine Learning Engineer at Clear AI
    Nov 2019 - Jun 2020 · 8 mos

    Principally worked on supply chain analysis, including optimising for multiple objectives and measuring the impact of extreme weather conditions.

  • Machine Learning Engineer at Argos
    Jun 2017 - Nov 2019 · 2 yrs 6 mos

    Recommender systems, including the product alternatives (which you can see under Or How About These) and the Get The Look for the Tu lines on Argos.

  • Data Science Engineer at Virtual Clarity
    Jan 2017 - Jun 2017 · 6 mos

  • Data Scientist at Royal Mail
    Dec 2015 - Dec 2016 · 1 yr 1 mo

    In this role, I worked principally on predicting delivery windows for parcels. Initially I joined an existing project, where my main role was to turn a Python prototype into a more scalable version by translating it into Scala and implementing some of the processes in Spark. Later, I created a prototype using a different approach that would let us take into account more of the variables we discovered impacted delivery times. This required collating data from a variety of sources, and getting a good idea of the inconsistencies between them as well as their own issues. It also required providing reports on progress to feed back to Operations on a daily basis, as well as provide information to senior stakeholders. I have also worked on a range of smaller projects, including: applying business rules alongside product recommendations, and detecting delivery points receiving regular high quantities of mail based on differing rule sets for different purposes. The main languages I used in this role were Python and Scala, using Teradata SQL to access data. I also worked somewhat with Excel to provide reports with detailed data on various projects.