Data Scientist

DATA SCIENTIST

Start - ASAP

Duration - 2 to 3 months

HYBRID - 2 days onsite / 3 days remote

Location - Chancery Lane, London

Daily rate - TBC

THE ROLE

We are seeking an exceptional Data Scientist with expertise in causal inference, experimental design, and conformal prediction to join our innovative analytics data science team.

In this role, you'll leverage advanced statistical methods to extract meaningful insights from complex data, design robust experiments, and develop predictive models with reliable uncertainty quantification.

Core Responsibilities

• Design, implement, and analyse causal inference experiments including natural experiments, and quasi-experimental methods
• Develop and apply conformal prediction frameworks to provide reliable uncertainty estimates for machine learning models
• Identify and control for confounding variables in observational studies
• Create robust statistical methodologies for causal effect estimation
• Collaborate with cross-functional teams to translate business questions into rigorous experimental designs
• Present technical findings to stakeholders in clear, actionable terms

Qualifications

• Advanced degree (MS or PhD) in a quantitative discipline with deep understanding of statistics
• 3+ years of professional experience in applying statistical methods to real data .
• Demonstrated expertise in experimental design, including randomized controlled trials and observational study methodologies
• Strong understanding of conformal prediction theory and applications
• Proficiency in programming languages such as Python or R, and relevant statistical packages
• Experience with causal inference frameworks (e.g., potential outcomes, causal graphs, do-calculus)
• Knowledge of modern machine learning techniques and how they intersect with causal reasoning
• Excellent communication skills, with ability to explain complex statistical concepts to non-technical audiences

Preferred Skills

• Experience with heterogeneous treatment effect estimation
• Familiarity with Bayesian methods for causal inference
• Background in epidemiology would be a plus
• Experience working with a causal inference ecosystem (pywhy, causal impact, synth, geolift,…)