Data Science

Angel•Provo, UT

3d•$110,000 - $130,000•Hybrid

About The Position

Angel Studios is transforming the entertainment industry by empowering its 2 million guild members to decide what content is produced and distributed. This role is crucial in understanding user engagement and behavior, building predictive models, and establishing the data foundation for personalized discovery. The data scientist will turn data into actionable insights, influencing how users discover and engage with Angel's growing library of stories. This position is the first dedicated data scientist for the Discovery team, focusing on the analytical foundation of the recommendation system. While the initial focus is on analytical rigor (metrics, experimentation, causal inference), there is a clear path to owning recommendation models in production.

Requirements

Statistical rigor: ability to design experiments correctly (power analysis, multiple comparisons, confidence intervals, Bayesian methods) and explain results to non-technical stakeholders.
Causal inference skills: experience with observational data, propensity score matching, difference-in-differences, instrumental variables, or regression discontinuity.
SQL and Python fluency for data exploration, analysis, modeling, and automation.
Experience designing, running, and analyzing A/B tests in production, understanding interaction effects, novelty effects, and Simpson's paradox.
Strong communication skills to translate complex analysis into clear narratives.
Data modeling experience with dbt or equivalent transformation frameworks.
Experience with large-scale user engagement and behavior data.
Track record of defining metrics frameworks that stakeholders adopted.
Familiarity with modern data tools: dbt, data warehousing (Snowflake, BigQuery, Redshift), experimentation platforms (GrowthBook, Optimizely), BI tools (Rill, Looker).
Must be authorized to work in the United States.

Nice To Haves

Python proficiency like a software engineer (tests, packaging, code reviews).
Experience with data pipelines, feature generation, monitoring, and retraining for deployed models.
Curiosity about systems design for ML features (latency, throughput, failure modes, observability).
Experience with any part of the lifecycle around a deployed model.
Experience with recommendation systems or personalization.
Familiarity with AI.

Responsibilities

Define, instrument, and maintain the Discovery metrics framework across web, mobile, and TV, including metrics like precision, recall, coverage, diversity, CTR, playthrough, completion, session depth, cold-start ramp time, and retention segmented by recommendation engagement.
Own the A/B testing and experimentation pipeline for Discovery surfaces, designing experiments with statistical rigor (sample sizing, duration, segmentation, guard-rail metrics).
Analyze user behavior across platforms (TV, mobile, web) to understand discovery, browsing, and engagement patterns, identifying insights related to Guild voting, theatrical-to-streaming conversion, content affinity, and churn risk.
Apply causal inference techniques to distinguish correlation from causation in engagement data, designing quasi-experiments when randomization is not feasible.
Build and maintain dbt models, data pipelines, and analytical infrastructure to ensure data accessibility and trustworthiness for the Discovery team and the broader organization.
Evaluate new signals (voting history, explicit ratings, content metadata, theatrical engagement) to improve recommendation recipe performance.
Prototype recommendation approaches (content-based filtering, hybrid models, embeddings) and evaluate them.
Take a model from notebook to production, including writing testable Python, managing data lifecycles (pipelines, feature stores, monitoring, retraining), and considering systems design (latency, failure modes, observability).