A year in the life of a Data Scientist at HMRC

When I found out that my first posting as a GSS Fast Streamer was as a data scientist in HMRC I had absolutely no idea what to expect.  The term data science was alien to me and how this differed from any other analyst role was not apparent.

Having now spent a year in the DS&T (Data Science and Technology) team within HMRC’s Analytical directorate KAI (Knowledge, Analysis and Intelligence), I am now a fully-fledged Data Science advocate. And, yes, I am still drowning in acronyms.

I have worked on a number of projects relating to self-assessment data, movements of duty-suspended goods around Europe, and have even explored machine learning techniques which could improve predictive modelling in HMRC.  Most excitingly, I have helped build a visualisation tool using the Java Script library D3 that will be used to identify and target large-scale alcohol fraud.  Working with an EU-wide database of commercial movements we have sought to maximise the benefits gained from existing data sources. This involved getting to grips with the basics of HTML/JavaScript/CSS whilst working with case teams on how such a tool could aid delivery.  Therefore this project has allowed me to develop technical programming skills as well as customer engagements skills through requirements gathering.

Coming straight from a Social Research MSc I was worried going into the role that I wouldn’t have all the skills required to carry out this range of work. However, whilst it has been a steep learning curve, I have been supported by my work team and the scheme with plugging my knowledge gaps. This has taken the form of training courses as well as peer support and knowledge sharing sessions across the KAI directorate.

The expectation in the Data Science & Technology team is that we are always looking for better ways to utilise data and are therefore constantly learning new things.  As the term ‘data science’ is used in many contexts (often in vague and confusing ways), it is this emphasis on innovation which has made my role feel ‘data science’ driven. I plan to try and bring this enthusiasm for new technologies/analysis techniques with me into all my future posts.  Anyone can be a Data Scientist!

Leave a comment

We only ask for your email address so we know you're a real person