P We’re hiring a strong Python‑first, statistics‑strong /strong Data Science Intern (or junior ML Engineer) to partner with our strong Data Analysts /strong on the hardest problems we have: strong probabilistic LTV /strong, strong churn prediction /strong, and strong forecasting /strong that actually moves CAC, payback, and ROAS. /p pstrong⚠️ READ THIS FIRST — Apply only if you meet this: /strong /pullistrong Eligibility: You’re currently studying at a university in Spain (ideally in or near Barcelona) and can sign a "convenio de prácticas". This position is open exclusively for internship agreements. /strong /li /ulp If you’re not sure whether your university can provide one, strong please confirm with them before applying /strong, as we’ll strong automatically filter out candidates who can’t meet this requirement. /strong /pp Otherwise, feel free to check out our other strong junior roles /strong on the Carrots Lab LinkedIn page › Jobs /pp where a "convenio" isn’t required. /p pstrong⭐ WHAT YOU’LL OWN (Your Mission) /strong /pullistrong LTV Prediction Probabilistic Forecasts /strong /lili Build, compare, and iterate LTV models (e.g., survival/retention‑based, zero‑inflated + Gamma, GBMs). Produce strong calibrated /strong predictions we can trust for UA and pricing. /lilistrong Churn Retention Modeling /strong /lili Predict churn windows and retention curves by app/geo/cohort to inform payback and creative/keyword strategy. /lilistrong Signal Engineering for UA /strong /lili Translate model outputs into strong conversion signals /strong (e.g., value‑based bidding targets) that improve Google/Meta optimization. /lilistrong Revenue Cohort Analytics (Python) /strong /lili Pull, transform, and join event purchase data (RevenueCat, AppsFlyer, Firebase, BigQuery) into model‑ready datasets. /lilistrong Decision Science /strong /lili Evaluate A/B tests (frequentist or Bayesian), pricing experiments, and funnel changes with strong confidence intervals /strong and strong practical uplift /strong. /li /ul pstrong WHAT SUCCESS LOOKS LIKE (90 Days) /strong /pullistrong LTV v1 shipped /strong per flagship app with baseline vs. model comparison (e.g., MAE/RMSE or calibration plots) and a documented retrain cadence. /lilistrong Churn or retention model v1 /strong that beats naive benchmarks (e.g., 30‑day retention mean) and informs UA payback assumptions. /lilistrong Signal spec /strong delivered: which in‑app events/values we should pass to ad platforms, how to weight them, expected CPA impact. /lilistrong Reproducible codebase /strong : clean notebooks/scripts with README, functions, and basic tests; data pulls automated in Python. /lilistrong Live dashboards /strong (with Sergi): LTV/payback cohort views wired to strong Looker /strong for weekly decision‑making. /li /ul pstrong /p