About the role
We are looking for two experienced Data Scientist / ML Engineer (4+ years) to build intelligent decisioning and AI systems for complex financial use cases. The role involves developing predictive models, next-best-action systems and agentic AI capabilities that translate customer and behavioural data into actionable decisions. You will own ML solutions end-to-end - from problem formulation and feature engineering to modelling, validation, deployment and monitoring - while working closely with the AI Product Lead and engineering teams.
Must-Have Skills:
- Strong proficiency in Python and SQL, with hands-on experience in Scikit-learn, XGBoost/LightGBM and TensorFlow/PyTorch.
- Strong experience building predictive and decisioning models using behavioural, transactional, bureau and other customer data.
- Experience with models/use cases such as Next Best Action, propensity modelling, payment/bounce prediction, customer segmentation, risk or response prediction.
- Hands-on experience taking ML models to production, including validation, monitoring, performance analysis and retraining.
- Strong statistical and modelling discipline, with the ability to identify data leakage, instability, bias and genuine predictive lift.
- Ability to translate ambiguous business problems into end-to-end ML/AI solutions and communicate trade-offs clearly to product, business and engineering teams.
- Working understanding of LLMs and Agentic AI, with the ability to integrate predictive ML models into larger AI/agentic decisioning workflows.
Preferred Skills:
- Experience with Next Best Action / recommendation / optimization / propensity modelling.
- Exposure to LLMs, agentic workflows, RAG or AI-driven decision systems.
- Working knowledge of AWS/GCP for ML and data workloads.
- Experience with MLOps - Git/DVC, experiment tracking, CI/CD, model monitoring and automated retraining.
- Sound understanding of data architecture, feature pipelines and low-latency model serving.
Requirements
- Strong proficiency in Python and SQL, with hands-on experience in Scikit-learn, XGBoost/LightGBM and TensorFlow/PyTorch.
- Strong experience building predictive and decisioning models using behavioural, transactional, bureau and other customer data.
- Experience with models/use cases such as Next Best Action, propensity modelling, payment/bounce prediction, customer segmentation, risk or response prediction.