Senior Data Scientist

LinkGroup is a global technology organisation.

• 5+ years of professional experience in Data Science, Machine Learning or related quantitative fields.
• Proven track record of deploying machine learning models to production.
• Strong Python skills and experience with platforms such as Databricks, Spark, and MLflow.
• Expertise in personalisation and recommendation systems.
• Hands-on experience with reinforcement learning and/or contextual bandits including Thompson Sampling and epsilon-greedy approaches.
• Strong understanding of experimentation and causal inference, including controlled experiments in high traffic environments.
• Experience with classical machine learning techniques like regression, classification, clustering, and recommendation.
• Experience with customer behavior data, segmentation, and production feature pipelines.
• Strong statistical reasoning and accurate evaluation of model and experimental results.
• Excellent communication skills for technical and non-technical stakeholders.
• Pragmatic, curious, and collaborative problem-solving approach.

Nice to have:

• Experience with off-policy evaluation, counterfactual learning, or propensity methods.
• Knowledge of sequential testing including SPRT.
• Experience with experimentation platforms such as Statsig or similar.
• Experience with streaming systems and low-latency model inference.
• Experience using unstructured data including embeddings and text signals.
• Background in marketing, CRM decisioning, customer communications, or funnel/conversion optimisation.

• Build, own and continuously improve ML and reinforcement learning models running in production.
• Develop decisioning logic for best next action across customer journeys and communication lifecycles.
• Design and evolve contextual bandit systems, including Thompson Sampling, offline policy evaluation, and low-latency deterministic inference.
• Collaborate with Engineering teams on system architecture and integration of models into production platforms.
• Define how models make decisions, serve decisions at scale, and safely handle failures and rollbacks.
• Lead experimentation for decisioning models, defining success metrics, guardrails, and methodologies.
• Apply statistical methods such as sequential testing, propensity methods, and causal inference to evaluate model performance.
• Monitor production models for performance, data quality, drift, and latency.
• Analyze customer behavior and translate model outputs for business and marketing stakeholders.
• Collaborate with Engineering, Marketing Analytics, CRM, and Product teams to produce measurable business outcomes.