Senior Associate - AI / Data

Department: IT

Employment Type: Permanent - Full Time

Location: Pune

Description

Key Responsibilities

• Assemble, cleanse, and model large, complex datasets from across the Group to meet functional and non-functional business requirements, using the medallion architecture on the Davies Microsoft data platform (Azure Data Factory, Microsoft Fabric, and Power BI).
• Design, build, and validate statistical and machine learning models for business-focused use cases including claims triage, fraud detection, outcome prediction, and operational efficiency optimisation.
• Collect and structure data through a combination of business analysis, study design, and automated collection pipelines — identifying the most valuable signals within large, complex datasets.
• Build and maintain analytical tools and reporting solutions that provide stakeholders with actionable insight into key business performance metrics, including operational efficiency, claims outcomes, and customer experience.
• Create clear, compelling reports, dashboards, and presentations that translate complex data findings into simple, actionable recommendations for business, product, and executive audiences.
• Develop and implement experimental frameworks for data collection and model evaluation, including A/B testing, model validation, and performance monitoring in production environments.
• Partner with data engineers, AI engineers, and platform teams to ensure data science outputs are production-ready, well-documented, and integrated effectively into Davies systems and workflows.
• Work with stakeholders across data, product, design, and executive teams to understand their data science needs, define problem statements clearly, and deliver solutions that are technically rigorous and practically useful. 
• Contribute to the continuous improvement of data science practice at Davies, including tooling, methodology, model governance, and knowledge sharing across the AI Office and wider Group.

Skills, Knowledge & Expertise

• Proficiency in Python and/or R for data analysis, statistical modelling, and machine learning, with strong command of relevant libraries (pandas, scikit-learn, PyTorch, or equivalent).
• Hands-on experience with the Microsoft data platform, including Azure Data Factory, Microsoft Fabric, and Power BI, for data ingestion, transformation, and visualisation.
• Experience designing and evaluating machine learning models — including classification, regression, clustering, and NLP techniques — and deploying them in production or near-production environments.
• Strong SQL skills and experience working with relational databases, data warehouses, and medallion architecture data lake patterns.
• Ability to visualise and communicate complex data in the most effective way possible — matching chart type, narrative, and level of detail to the audience and business question.
• Excellent written and verbal communication skills, demonstrated through the preparation of reports, presentations, and technical documentation for diverse audiences including senior leadership.
• Analytical and problem-solving mindset with exceptional attention to detail — comfortable searching through large, messy datasets to surface usable and statistically meaningful insight.
• Ability to work independently and collaboratively across teams with different technical backgrounds, translating between data science concepts and business requirements with clarity.
• Experience building data transformation processes that handle workload management, data structures, dependencies, and metadata in a governed, repeatable way.
• Familiarity with model monitoring, experiment tracking, and MLOps principles — ensuring that models remain performant and auditable over time.
• Continuous improvement mindset, naturally inquisitive and intellectually driven, with a passion for finding signal in data and translating it into business impact.
• Experience applying data science in insurance, financial services, claims management, or other regulated industries is a strong advantage, particularly where data governance and model explainability are key requirements. 
• Familiarity with AI and generative AI tooling — including LLMs, embedding models, and vector search — and an interest in how classical data science and modern AI methods can be combined to deliver greater business value is an advantage.