Research Engineer - Contextual Bandits & RL
UK Visa Sponsorship Analytics
Analytics are greyed out due to low classification confidence (37.0%).
Occupation TypeTelecoms and related network installers and repairers
Occupation Code Skill LevelMedium Skilled
Sponsorship Salary Threshold
£41,700
(£21.38 per hour)
Standard minimum applies
Above analytics are generated algorithmically based on job titles and may not always be the same as the company's job classification. You can also check detailed occupation eligibility, and salary criteria on our UK Visa Eligible Occupations & Salary Thresholds page.
Disclaimer: Hunt UK Visa Sponsors aggregates job listings from publicly available sources, such as search engines, to assist with your job hunting. We do not claim affiliation with algo1. For the most up-to-date job details, please visit the official website by clicking "Apply Now."
Description
About Us
We are a VC-backed startup focused on hyper-personalisation, currently in stealth. Inspired by the latest in recommender systems, we leverage transformers and graph learning alongside decision-making models to build the most engaging customer experiences for in-store retail.
Our mission is to change retail forever through hyper-personalised experiences that are both simple and beautiful.
About the Role - Offline Contextual Bandits and RL for Hyper-personalisation
We are looking for a Research Engineer to build decision-making models for in-store hyper-personalisation, with an initial focus on learning from logged human interaction data in an offline setting. You will work closely with domain experts and engineers to develop contextual bandit and reinforcement learning approaches that can support both single-step decisions and multi-step customer journeys, with the potential to enable online learning over time.
Key Responsibilities
We are a VC-backed startup focused on hyper-personalisation, currently in stealth. Inspired by the latest in recommender systems, we leverage transformers and graph learning alongside decision-making models to build the most engaging customer experiences for in-store retail.
Our mission is to change retail forever through hyper-personalised experiences that are both simple and beautiful.
About the Role - Offline Contextual Bandits and RL for Hyper-personalisation
We are looking for a Research Engineer to build decision-making models for in-store hyper-personalisation, with an initial focus on learning from logged human interaction data in an offline setting. You will work closely with domain experts and engineers to develop contextual bandit and reinforcement learning approaches that can support both single-step decisions and multi-step customer journeys, with the potential to enable online learning over time.
Key Responsibilities
- Develop and productionise offline contextual bandit and offline RL methods that learn from logged interaction data.
- Build rigorous off-policy evaluation (OPE) and counterfactual validation to measure candidate policies offline and compare approaches reliably.
- Formulate and model both single-step decisions (contextual bandits) and multi-step decision processes (sequential / RL style settings) based on real retail interactions.
- Advance representation learning for decision-making, including using transformers and GNNs where appropriate for behavioural, relational, and sequential data.
- Translate research ideas into robust systems: dataset design, modelling, evaluation, deployment, monitoring, and iteration.
- Collaborate cross-functionally to turn ambiguous product goals into concrete ML objectives, experiments, and deliverables.
- 3 to 5+ years applying machine learning research in production settings.
- Demonstrated experience with offline policy learning and evaluation methods (for example IPS style estimators and doubly-robust approaches, plus uncertainty estimation).
- Familiarity with bandit algorithms and exploration strategies, with interest in enabling online learning when the product is ready.
- Experience with recommenders and ranking (candidate generation, reranking, slates).
- Experience building data pipelines and improving data quality in modern ML environments.
- PhD in a relevant field.
- Opportunity to build technology that will transform millions of shopping experiences.
- Real ownership and impact in shaping product and company direction.
- A dynamic, collaborative work environment with cutting-edge ML challenges.
- Competitive compensation and equity in a rapidly growing company.