> Markdown version of [/jobs/ext/4031523-staff-decision-scientist-advanced-analytics](https://www.wearedevelopers.com/jobs/ext/4031523-staff-decision-scientist-advanced-analytics). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # Staff Decision Scientist, Advanced Analytics - **Company:** Life360 - **Location:** Beverly Hills, CA, United States (Remote available) - **Experience:** Expert - **Salary:** $130,500.0 - $242,500.0 - **Contract:** Permanent contract - **Skills:** Artificial Intelligence, Data Analysis, Code Review, Global Positioning Systems (GPS), GPS Tracking Unit, Mobile Application Software, Python (Programming Language), SQL Databases, Claude Code, Statsig, Virtual Agents, Optimizely - **Published:** October 30, 2026 - **Apply:** https://www.careerbuilder.com/job-details/staff-decision-scientist-advanced-analytics-beverly-hills-ca--429aafab-e2c6-4a15-8fc8-52b29dcd8247 ## About the Role Mobile Applications, GPS (Global Positioning System), Work From Home, Artificial Intelligence (AI), DNA, Organizational Skills, Data Science, Data Analysis, Product Marketing, Customer/Consumer Behavior, * Methodological depth with judgment: You know the assumptions behind each method, when they break, and when a simpler approach is the better answer. * Ownership mentality: You own the forecast and the standards end to end, including tracking whether they hold up and fixing them when they don''t. * AI-native working style: You use AI tooling (Claude Code or equivalent) as a genuine development partner: delegating discrete tasks, reviewing outputs critically, and running parallel workstreams. You think about how AI can scale the work of the whole team, not only your own. * Multiplier mindset: You measure your success by how much better the team''s work gets. You enjoy teaching and you''re patient with people learning new methods. * Influence without authority: You''ve driven adoption of standards and methods across teams you don''t manage by making the case clearly and making the right way the easy way., * 8+ years in an analytics, data science, or decision science role at a consumer tech company. * Advanced degree in a quantitative field (economics, statistics, quantitative social science, operations research) or equivalent practical experience. * Deep, applied experience with causal inference methods (e.g., difference-in-differences, instrumental variables, regression discontinuity, synthetic controls, propensity score matching). * Experience building and owning a forecast or growth model for a core business metric (e.g., MAU, DAU, subscribers, revenue), including cohort- or component-based approaches. * Expertise in experiment design and analysis at scale, including variance reduction, sequential testing, and network effects or interference. * Experience with experimentation platforms (Statsig, Optimizely, or similar). * Track record of setting analytical standards and mentoring other data scientists. * Proficiency in SQL and Python/R for statistical analysis. Preferred Qualifications * Experience with subscription or freemium business models. * Background in growth, retention, or lifecycle analytics. * Experience with Bayesian methods or probabilistic programming (e.g., PyMC, Stan). * Experience with LTV modeling, incrementality testing, or marketing mix modeling. * Experience building internal tooling or AI agent workflows for analytics teams. * Experience building dbt models or contributing to analytics engineering workflows. * Familiarity with international / multi-market analytics. ## Description Life360's mission is to keep people close to the ones they love. Our category-leading mobile app, Tile tracking devices, and Pet GPS tracker empower members to protect the people, pets, and things they care about most with a range of services, including location sharing, safe driver reports, and crash detection with emergency dispatch. Life360 serves approximately 102.4 million monthly active users (MAU), as of June 30, 2026, across more than 180 countries., * Own the MAU forecast and growth model. Build and maintain the model (Python, SQL) that breaks MAU into registration, retention, and reactivation, and that leadership relies on for planning and target-setting. Make the assumptions explicit, track forecast accuracy, and explain the misses. * Raise the bar on experimentation. Set the standard for experiment design and analysis: power, metric selection, variance reduction, sequential testing, heterogeneous effects, and interference in a product where users share Circles with family and friends. Be the reviewer the team seeks out for high-stakes tests. * Push our causal inference toolkit forward. When a clean experiment isn''t possible, bring the right quasi-experimental method (difference-in-differences, synthetic controls, regression discontinuity, instrumental variables, and others) and turn one-off analyses into reusable approaches others can apply. * Set the methodological standard for the data team. Define and document best practices, review analyses across OGs, and be the go-to person for hard methods questions. * Collaborate across disciplines. Work with OG-embedded Decision Scientists on their toughest problems, with Data Scientists on modeling approaches, and with Analytics Engineers and Data Engineers to turn methods into durable data models, metrics, and tooling. * Scale advanced analytics through AI. Build agent workflows (Claude Code or equivalent), templates, and tools that let the team run forecasts, experiment readouts, and causal analyses faster and more consistently. Treat every repeated analysis as a candidate for automation. * Teach and mentor. Level up the craft across the org through mentorship, analysis and code reviews, internal talks, and written guides. Raise the ceiling for senior ICs and the floor for everyone else. * Make complex methods land with leadership. Translate forecasts and causal findings into clear recommendations with a point of view. A rigorous result nobody acts on is a failed result.