Data Science · London / Remote · Full-time

Biostream; Greater London; Hybrid; GBP 75,000 - 110,000; Full time

Job summary

Biostream is seeking a biostatistician with strong Bayesian modelling skills to develop probabilistic approaches for physiological data and medical risk. The role involves integrating data sources to estimate deterioration probabilities and to support triage prioritisation in real-world, noisy settings.

You will collaborate with AI engineers, medical advisors, and product teams to design models that operate in austere environments and handle incomplete data, contributing to publications and

Qualifications

• PhD or MSc in biostatistics, epidemiology, statistics, applied mathematics, or a related field.
• Experience with Bayesian inference and probabilistic modelling.
• Familiarity with health data, physiological signals, or medical datasets.
• Strong programming skills in Python or R.

Responsibilities

• Develop Bayesian statistical models to estimate physiological deterioration risk from multi-sensor data
• Build probabilistic frameworks integrating vital signs, haemodynamic indicators, motion data, and environmental variables
• Design and run simulation studies to test triage algorithms and risk scoring systems
• Work on uncertainty quantification and probabilistic inference for medical decision support
• Collaborate with engineers to integrate statistical models into machine learning pipelines
• Analyse physiological and operational datasets to identify predictive patterns
• Support validation of models against real-world medical outcomes
• Contribute to scientific publications and technical documentation

Skills

Bayesian inference

Probabilistic modelling

Python

R

Data analysis

Education

PhD in Biostatistics

MSc in Biostatistics or related field

Tools

PyMC

Stan

TensorFlow Probability

Job description

Role

We are looking for a biostatistician with expertise in Bayesian modelling to work on probabilistic modelling of physiological data and medical risk.

The role focuses on developing statistical frameworks that combine multiple data sources to estimate the probability of life-threatening deterioration and support triage prioritisation.

You will work closely with AI engineers, medical advisors, and the product team to design models that operate in real-world environments with noisy and incomplete data.

Responsibilities

• Develop Bayesian statistical models to estimate physiological deterioration risk from multi-sensor data
• Build probabilistic frameworks integrating vital signs, haemodynamic indicators, motion data, and environmental variables
• Design and run simulation studies to test triage algorithms and risk scoring systems
• Work on uncertainty quantification and probabilistic inference for medical decision support
• Collaborate with engineers to integrate statistical models into machine learning pipelines
• Analyse physiological and operational datasets to identify predictive patterns
• Support validation of models against real-world medical outcomes
• Contribute to scientific publications and technical documentation

Profile

• PhD or strong MSc in biostatistics, epidemiology, statistics, applied mathematics, or a related field
• Experience with Bayesian inference and probabilistic modelling
• Familiarity with health data, physiological signals, or medical datasets
• Strong programming skills in Python or R
• Experience with probabilistic frameworks such as PyMC, Stan, TensorFlow Probability, or similar
• Ability to work with noisy or incomplete real-world data

Nice to Have

• Experience with physiological monitoring, biosensors, or wearable devices
• Experience with causal inference or survival analysis
• Experience working with clinical datasets or emergency medicine
• Interest in defence technology, emergency response, or austere medical environments