Data Scientist

Design, build, and evaluate classical machine learning models for business-critical use cases (classification, regression, ranking, anomaly detection, time-series forecasting).

Apply probabilistic and Bayesian modeling techniques to quantify uncertainty and inform decision-making under uncertainty; leverage tools like PyMC and PyMC-Marketing for Bayesian workflows.

Perform rigorous EDA, feature engineering, and data wrangling on large structured and semi-structured datasets using Python and SQL.

Collaborate with data engineers and analytics engineers to source, clean, and validate data pipelines feeding ML workflows.

Develop, track, and communicate model performance metrics; identify degradation signals and recommend retraining or improvement strategies.

Translate business questions into well-framed statistical problems and present findings clearly to technical and non-technical stakeholders.

Maintain clean, reproducible, and well-documented code and notebooks following team engineering standards.

3–4 years of hands-on experience in a data science or applied ML role.

Strong command of classical ML algorithms - gradient boosting, random forests, SVMs, logistic regression, clustering, dimensionality reduction, etc.

scikit-learn, XGBoost, LightGBM, CatBoost. Proficiency with ML frameworks:

PyMC or PyMC-Marketing. Solid understanding of probabilistic modeling, Bayesian inference, and uncertainty quantification; working experience with

Python (pandas, NumPy, SciPy, matplotlib/seaborn/plotly, MLflow). High proficiency in

SQL skills - complex multi-table queries, window functions, performance optimization. Strong

Deep familiarity with model evaluation frameworks: cross-validation, calibration, AUC, RMSE, MAPE, lift/gain curves, and business-aligned metrics.