We’re building an analysis engine that helps organisations discover what is happening
and why. It makes use of the evidence conventional
data systems miss – from large internal text datasets to signals spread across
many external sources. Our architecture combines comparative statistical
inference and causal frameworks with machine learning and large language
models, ensuring stable interpretation and
consistent quality at scale.
We’re backed by a specialist AI venture fund, as well as receiving EU deep-tech funding.
Our team’s previous research has been cited over 35,000 times and has previously informed government and industry decision-making.
We’re now working with teams across government, pharma, and other large organisations,
scaling our first deployments and building the scientific foundation for
what comes next.
For too long, organisations have settled for a shallow, lagged view of the dynamics they care about. We’re changing that,
helping teams uncover the drivers that matter and build understanding that
improves over time. This is a hard,
meaningful problem – come help us solve it.
We’re hiring an Applied Scientist to invent, validate and implement methods for
working out what drives outcomes from unstructured evidence. You’ll explore the
frontier between statistical and causal inference and large language models – designing
experiments, prototyping pipelines, and collaborating with engineers to
bring your best ideas into production quickly.
You’ll work directly with our CTO – a mathematician with deep experience
of turning complex data into real-world impact – and collaborate with
enterprise clients to ensure our analysis is rigorous, scalable, and
relevant to decision-makers. As one of our early hires, you’ll help shape Wholesum’s
methodology, culture, and future.
You’re excited by the challenge of understanding cause and effect in messy, real-world data, and have a strong background spanning statistical inference, causal inference and machine learning.
You’re excellent at pushing methodological boundaries and rapidly turning insights into well-designed Python pipelines.
You spend a lot of your time innovating with LLMs, moving beyond basic tools, and enjoy working out where models fail and why.
You have substantial experience evaluating and benchmarking statistical, ML and/or AI outputs.
Experience producing analysis that informed real decisions under scrutiny and time pressure.
Interest in human judgement, behaviour and decision-making.
Understanding of common data collection and sampling methods.
Experience building in early-stage or startup environments.
Form not loading? Open it in a new tab.