← Role tracks

R4 Stage D · 160 h (56 T / 104 P)

Data Scientist

Converts business questions into measured, defensible, decision-grade analysis and models.

Tech stack

PythonSQLscikit-learnXGBooststatsmodelsSHAPProphetPlotlyStreamlitPower BI/Tableau

Modules

M1 · Problem framing and metric design

36 h
TOPICS
  • Translating business objectives into measurable targets
  • Primary versus guardrail metrics
  • Proxy-metric risk
  • Baselines
  • Cost of errors
  • Decision thresholds
  • Stakeholder alignment

Lab: Convert an ambiguous business objective into a measurable analytical plan

Course material for this module is in production.

M2 · Inference and experimentation

44 h
TOPICS
  • Sampling and estimation
  • Confidence intervals
  • Hypothesis testing
  • Multiple comparisons
  • Power and sample-size calculation
  • A/B test design and analysis
  • Sequential testing pitfalls
  • Quasi-experiments
  • Confounding and causal reasoning

Lab: Design, power and analyse a controlled experiment; write the decision memo

Course material for this module is in production.

M3 · Advanced modelling and interpretability

44 h
TOPICS
  • Regularisation
  • Ensembles
  • Time-series forecasting and seasonality
  • Survival and uplift concepts
  • Calibration
  • Feature attribution with SHAP
  • Partial dependence
  • Fairness assessment
  • Model limitations

Lab: Build a calibrated, explained model with a documented limitations section

Course material for this module is in production.

M4 · Communication and decision support

36 h
TOPICS
  • Narrative structure
  • Executive summaries
  • Visual encoding principles
  • Dashboard design
  • Uncertainty communication
  • Recommendation framing
  • Handling challenge from stakeholders

Lab: Present findings and defend the recommendation to a challenging panel

Course material for this module is in production.

Track project

End-to-end business analytics engagement: framing, data preparation, calibrated model with interpretability, decision dashboard, quantified recommendation with expected impact and uncertainty, delivered as a stakeholder presentation.

Job-ready exit standard

Frames a business problem, models it and communicates a decision; SQL fluency assumed.