You will own the full lifecycle of data science solutionsfrom framing business problems and exploring raw data to developing, deploying, and continuously improving models in production.

The role requires strong statistical thinking, hands-on ML skills, and the ability to convert complex business challenges into scalable data products.

You will collaborate closely with product, engineering, and domain specialists, particularly in the insurance and commercial data ecosystems.

Responsibilities -:

- Translate open-ended business questions into clear statistical/ML problem statements, hypotheses, and measurable success metrics.

- Explore and profile datasets to understand distributions, correlations, data quality issues, and feature opportunities.

- Design, train, validate, and compare statistical and machine learning models (regression, classification, forecasting, uplift modeling, etc.) with strong attention to baselines and evaluation rigor.

- Own end-to-end model lifecycle : experimentation, model selection, packaging, deployment, monitoring, and continuous improvement.

- Design statistically robust experiments (A/B tests, hypothesis tests, quasi-experiments) to quantify impact and support product decisions.

- Collaborate with data engineering and backend teams to integrate models into real-time or batch production pipelines.

- Build reusable components feature pipelines, validation/evaluation scripts, monitoring checksto accelerate team delivery.

- Analyze production performance, identify model/data drift, and lead corrective actions for reliability and robustness.

- Present analytical findings and model insights to both technical and non-technical stakeholders using documents, dashboards, and presentations.

- Provide guidance to junior data scientists on statistical methods, modeling best practices, and coding standards.

Must-Have Skills :

- Strong programming skills in Python, including pandas, NumPy, scikit-learn; ability to write clean, modular, well-tested code.

- Solid command of statistics: hypothesis testing, confidence intervals, distributions, sampling, experiment design, regression analysis, time-based effects, and bias/variance concepts.

- Good working knowledge of SQL and experience handling structured datasets.

- Hands-on experience developing and deploying ML models from scratch into production (beyond POCs/notebooks).

- Experience analyzing model performance, doing error analysis, and diagnosing drift or performance degradation.

- Strong problem-solving skills, numerical aptitude, and comfort working with real-world, imperfect data.

- Ability to communicate complex statistical and ML concepts effectively to cross-functional teams.

- Self-driven and adaptableable to manage ambiguity, prioritize effectively, and take ownership end-to-end.

Good to Have :

- Prior experience working in the insurance domain (commercial or personal lines) or adjacent data-heavy domains.

- Exposure to working with US or UK stakeholders/clients, with comfort in cross-time-zone collaboration.

- Willingness to align partial working hours with US-based teams when required.

- Experience with MLOps tooling: Docker, MLflow, model serving APIs, CI/CD for ML, monitoring frameworks.

- Experience with LLM-enabled workflows (RAG, prompt engineering, LLM evaluation).

- Familiarity with dashboarding/visualisation tools such as Streamlit, Plotly, Power BI, or similar.

- Experience working in cloud environments (AWS, Azure, or GCP).

Education & Experience :

- Undergraduate or graduate degree in fields such as Statistics, Mathematics, Computer Science, Data Science, Economics, or similar quantitative disciplines.

- 3 to 5 years of professional experience in applied data science or ML, including ownership of at least one end-to-end production model or analytics solution.