We are the DataLab, at the University of Oxford, led by Dr Ben Goldacre. We are a small, productive, very fast-moving team of software developers, academics, and clinicians, all pooling skills and knowledge to build platforms and tools as well as traditional academic papers. We build software that has a real impact on the world, including: OpenSAFELY.org (the largest primary care data research platform in history, with 24 million patients’ data); OpenPrescribing.net (with 130,000 unique users a year); and TrialsTracker.net (which has changed policy and practice on clinical trial reporting). We are seeking new engineers to join our tight, fun, fast-moving team.
You’ll design and build intuitive, open source, web-based analysis tools, incorporating our innovative data science techniques, that work alongside the papers we write.
You’ll collaborate on high-impact academic papers, which reference the tools we build.
You’ll help the team use these to campaign for better use of data and software in science and medicine.
Our work is influential: it features in high-profile mainstream media stories, drives the agenda at parliamentary committees, and is published in a wide range of very high impact journals. We are currently focussing on building OpenSAFELY: an innovative, open source research environment which places patient privacy first. This is currently being used to analyse tens of millions of patient records to inform the global response to COVID-19. You can read a summary of press coverage on the OpenSAFELY website. We don’t just produce papers and tools, we advocate for better use of software, including our campaign for all research to share their code.
Candidates who are a good fit will:
- Be proficient with Python, Django, and SQL, and the usual Open Source tooling, e.g. git, code reviews, continuous integration
- Actively contribute to a supportive, helpful culture
- Have a natural curiosity about topics outside their normal area of expertise
- Be motivated by working very closely with end users
- Enjoy working independently, in a small team, with a light management touch
- Have an enthusiasm for solving problems in medical informatics
- Be happy switching between quite different domains – we take joint responsibility for all our outputs and activities
- Be happy taking responsibility for complex apps: this means new features, maintenance and devops
- Have a “leave it better than you found it” mentality, and are willing to work with and improve on code you did not originally write.
To give you a flavour of the range of activities you might work on, this is a selection of things we’ve worked on in the last year:
- Prototyping a new application for analysing variation in blood testing in primary care
- Building a secure environment for running analyses against confidential medical records
- Creating a novel column-oriented data infrastructure to support optimal real-time querying of a large prescribing dataset – and migrating existing functionality to use it
- Refactoring existing functionality
- Developing our custom slackbot for improved devops
- Many small fixes, such as writing new scrapers
- Teaching clinicians how to use GitHub, and learning from clinicians about medicine coding systems
- Writing a web app to support structured input of clinical coding data to our OpenSAFELY analytics platform
- Publishing a paper about health informatics in the NHS
We are looking for a developer to join our small team of three engineers. You will work across all our outputs, including the software listed above, and devops. This will be on a short- or long-term contract basis with the potential for longer-term employment in the future. We mostly work remotely; pre-pandemic, we met once or twice a month in Oxford, so you should be based in the UK. We are able to offer flexible working and a full-time or part-time post.
Our core stack is Python + Django + PostgreSQL, running in VMs. We farm out computationally expensive work to Google BigQuery (e.g. for complex queries that scan all 90GB of processed prescribing data). We generate charts using HighCharts. We are starting to use Docker more widely. We quickly prototype “data science” ideas in Jupyter Notebooks (example), and are able to generate tools and blog posts as rapid responses to media stories as they unfold (example). Everything we do is open by default.
We’re far more interested in practical experience, pragmatic problem solving skills, and the ability to work in a mixed team than we are in formal qualifications or big names on your CV. And we’d actively welcome applications from those who don’t have a traditional tech/academic background.
If you’re interested please send a CV, plus your expected day rate and ideal working hours, to email@example.com. We can offer £350 – £450 per day. We encourage experienced candidates who can earn more than this in other sectors to apply, so we can sell you the experience of working in such a productive team on such interesting and important work!
Your cover letter must clearly highlight evidence from your CV to explain why you are a good fit. Applications close on 24th June. Following initial screening, we will interview up to 10 candidates remotely. We will then provide a short (no more than 2 hours) coding task for the final two candidates, plus a series of very short interviews with three other team members.