Sr. Data Science Consultant (2-3 months)
Objective
Engage an external senior Data Science consultant to audit, redesign and rebuild our chargeback estimation and revenue estimation models, delivered as production-grade pipelines in our existing stack (dbt on Databricks, Python), plus a business-facing scenario tool that lets non-technical stakeholders input hypotheses (for example, price changes) and see the impact on expected revenue.
Methodology preference
We explicitly favour simple, interpretable, parametric approaches over machine learning black boxes:
Closed-form equations, demand/price curves, elasticity models, cohort survival curves, parametric distributions, GLMs, additive decompositions (trend × seasonality × price × mix), and Bayesian priors where they help with uncertainty.
Every parameter must have a business meaning the team can explain in plain language (if
price goes up 10%, conversion drops by X% because elasticity = -1.3).
Tree ensembles, deep learning and LLMs are out of scope unless the consultant can make a strong case that a parametric alternative is unworkable.
Bias toward fewer, well-understood features with documented assumptions, rather than high-dimensional models with opaque outputs.
Engagement shape
Type: fixed-scope consulting engagement, audit and rebuild.
Duration: approximately 2-3 months (10-12 weeks).
Dedication: full-time or near full-time (4-5 days/week) preferred. Part-time accepted if the duration is extended.
Modality: remote with occasional on-site / sync workshops. Same or adjacent timezone to Spain (CET ‡)
Required profile
5-8 years of senior data science / analytics engineering experience.
Has owned a forecasting or financial-estimation model in production end-to-end.
Bias toward parsimony: demonstrable preference for parametric / curve-based / closed-form approaches (price elasticity, demand curves, cohort/LTV curves, survival models, GLMs) over black-box ML when the problem allows it.
Has built scenario / what-if tools for non-technical stakeholders (price sensitivity, revenue planning, unit economics).
Stack: Python (pandas, numpy, scikit-learn, statsmodels and/or scipy.optimize), dbt (incremental models, tests, snapshots), Databricks (PySpark, Delta, jobs) and advanced SQL.
Domain: demonstrable work on chargebacks, payment risk, refund/dispute modelling, revenue forecasting, LTV, cohort revenue projections, or pricing/elasticity.
Languages: Spanish C1+ (working day-to-day with finance/ops stakeholders) and English
B2+ (documentation).
Comfortable reconciling model output against accounting figures and explaining uncertainty in business terms.
Nice to have
Payments / fintech / subscription / e-commerce background.
MLflow on Databricks, Unity Catalog, Databricks Workflows or Airflow.
Prior consulting / fixed-scope engagement track record (references).
Benefits we offer
Access to e-Learning platforms.
Amazing people-oriented organizational culture
Working from anywhere
Challenging projects using the latest technologies with clients from the US.