Senior Data Scientists Analytics Leads

Must Have Technical/Functional Skills:

• 10+ years in data science, advanced analytics, or applied statistics, including 3+ years leading analytical workstreams or mentoring data scientists.
• Must have personally built and validated predictive models; BI-only leadership is not sufficient.
• Strong Python and SQL, including pandas, NumPy, scikit-learn, and XGBoost or LightGBM.
• PySpark for larger datasets and experience with environments such as Databricks, SageMaker, or Vertex AI are useful, not required.
• Solid statistics: hypothesis testing, linear and logistic regression, time-series methods, sampling, and clear communication of uncertainty.
• Model development for at least one BFSI use case such as credit risk, fraud/AML alert prioritization, propensity/churn, collections, or forecasting.
• Choose validation and metrics appropriate to the problem (e.g., out-of-time tests, AUC/KS for scorecards, precision/recall for alerts, MAE for forecasts).
• Experience measuring business impact with A/B tests where feasible or with defensible observational approaches such as matched cohorts, difference-in-differences,
• or controlled pre/post comparisons; not every method is required.
• Explain ability and model governance: use SHAP or similar techniques, assess bias where relevant, and document assumptions, limitations, and validation.
• For US banking models, familiarity with SR 11-7 expectations is useful; credit-decision reason codes apply only to relevant credit use cases.
• Production awareness:
• Git, code review, reproducible pipelines, unit and data tests, MLflow tracking, and handoff to ML engineering for batch scoring or APIs, with monitoring such as PSI and performance drift.
• Clear presentation of results:
• KPI definitions, visuals, and decision rules or score cut-offs that business teams can act on.
• Power BI or Tableau is a plus.
• Practical GenAI use: applying LLMs to classify, summarize, or extract data from unstructured text (e.g., complaints, KYC documents) and evaluating output quality.
• LLM infrastructure experience is not required.
• Client-facing leadership:
• frame ambiguous problems, agree on outcomes, review code and results, and guide a small team; experience coordinating across time zones is useful when delivery is distributed.
• BFSI experience strongly preferred.

• Turn a client business question into an analysis plan covering target population, data sources, baseline, success metric, and milestones.
• Lead hands-on data exploration and preparation in SQL, Python, or PySpark; resolve data issues with data engineering and document definitions and exclusions.
• Build and review models for BFSI use cases, choosing validation schemes and metrics that reflect business cost (e.g., fraud losses vs. analyst review effort).
• Measure business impact through A/B tests or quasi-experimental methods, reporting uplift with confidence intervals.
• Deliver recommendations as presentations, dashboards, or score cut-offs and decision rules agreed with the business.
• Lead a small team of data scientists and analysts: break down work, review code and results, manage delivery quality, and contribute hands-on; coordinate onshore/offshore contributors where the engagement requires it.
• Hand off production-ready models to ML engineer ing with scoring specifications, test datasets, and monitoring thresholds; agree ownership for release and support.
• Track models after launch:
• PSI and drift, actual vs. expected performance, and business adoption; recommend recalibration or retraining when needed.
• Prepare model-development documentation and support independent validation where applicable, addressing findings with evidence such as validation results, SHAP explanations, or fairness tests; include credit-decision reason codes when relevant.
• Report progress, findings, and limitations to client sponsors, and help scope and estimate follow-on work.

• Strong leadership and stakeholder management skills with the ability to work effectively across business, data, engineering, and architecture teams.
• Client-facing ability, excellent communication and presentation skills for technical and business.
• Strong consulting and problem-solving mindset with the ability to evaluate options and recommend practical solutions.
• High ownership, accountability, and attention to detail with a focus on quality, governance, and customer success.
• Strong client-facing experience supporting workshops, architecture reviews, solution presentations, and executive discussions.