About the role
We're looking for a Senior Data Analyst / Data Science Engineer with 7+ years of handson experience in Python-based analytics, statistical modelling (regression, forecasting, GLM), and strong data wrangling. You will convert complex Excel models into Python, validate parity, enhance them with software engineering best practices, and deploy them as reusable components/pipelines. The ideal candidate is equally comfortable discussing model assumptions with business stakeholders and implementing productionready code with tests and documentation.
Responsibilities
- Migrate existing Excel-based analytical/financial/statistical models into Python with feature parity, improved performance, and robust error handling.
- Design, build, and validate regression and related statistical models (e.g., Linear/Multiple Regression, Logistic, Elastic Net, Timeseries/ARIMA/Prophet, Panel data as applicable).
- Data wrangling & EDA: Clean, join, transform large datasets; exploratory analysis; feature engineering; handling outliers and missing data.
- Model governance: Document assumptions, diagnostics (VIF, residuals, multicollinearity, heteroskedasticity), backtesting, and performance metrics (RMSE/MAE/AUC as relevant).
- Productionization: Package models into versioned Python modules/notebooks/pipelines; write unit tests, data validation checks, and automation scripts (e.g., cron/Airflow).
- Stakeholder engagement: Translate business questions into modelling problems; communicate insights, tradeoffs, and explain model outcomes to nontechnical audiences.
- Collaboration: Partner with data engineering to access data; with BI teams to publish results (dashboards, APIs); with product/ops to integrate into decision flows.
- Optimization: Refactor legacy logic, improve runtime, and reduce manual effort through automation and reproducible analysis.
- Documentation & handover: Clear readme, change logs, model cards, and SOPs for ongoing maintenance.
Musthave qualifications
- 710 years in analytics/data science with demonstrable Python proficiency (pandas, numpy, statsmodels, scikitlearn).
- Strong command of regression techniques (linear, logistic, regularization, model diagnostics, crossvalidation).
- Proven experience converting Excel models into Python (e.g., complex formulas, lookups, macros/VBA logic translated to Pythonic pipelines).
- SQL proficiency for data extraction and joins; solid EDA and data quality practices.
- Clear, structured communication; stakeholder management and requirement grooming.
- Version control (Git), notebook hygiene, and documentation habits.
Nice to have
- Timeseries modelling (ARIMA/SARIMA/Prophet), survival analysis, or Bayesian methods.
- Workflow orchestration (Airflow/Prefect), CI/CD for analytics, basic Docker.
- Experience with Excel/VBA for model understanding; OpenPyXL/xlwings helpful.
- Cloud exposure (AWS/GCP/Azure) for data/compute; basic API integration.
- Visualization: matplotlib/seaborn/plotly; BI tools (Power BI/Tableau).
- Domain exposure to financial modelling, pricing, marketing analytics, risk, or operations (as applicable).
