> Markdown version of [/jobs/ext/4758805-principal-data-scientist](https://www.wearedevelopers.com/jobs/ext/4758805-principal-data-scientist). 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). --- # Principal Data Scientist - **Company:** Self Financial - **Location:** Austin, TX, United States (Remote available) - **Experience:** Expert - **Salary:** $200,000 - $240,000 - **Contract:** Permanent contract - **Skills:** Amazon Web Services, Amazon S3, Code Review, Python (Programming Language), NumPy, SQL Databases, Feature Store, Feature Engineering, Large Language Models, Pandas, Production Code, Machine Learning Operations, Software Version Control - **Published:** November 17, 2026 - **Apply:** https://startup.jobs/principal-data-scientist-self-financial-10795675 ## About the Role * 8+ years of hands-on data science and machine learning, with models shipped to production and kept performing, tied to business outcomes. * Expert-level Python (Pandas, NumPy) and SQL. * Applied modeling experience across lead scoring, propensity, LTV, time-series forecasting, clustering, recommenders, and causal inference. * Experience with experimentation, uplift modeling, or personalization. * Ownership of the full model lifecycle: feature engineering on large, complex datasets, training, tuning, validation, and retraining in production, in partnership with MLOps and Engineering on deployment and monitoring. * A track record of setting technical standards that others adopted, and of influencing senior stakeholders. You can explain a model's tradeoffs to a VP in plain language. * Strong engineering practices, including version control, code review, and testing. * Comfort being first. You're fine without an established data science team and will help define what good looks like., * A Master's or PhD in a quantitative field (CS, Statistics, Physics, Math). * Fintech or financial services experience. * Experience with real-time (low-latency) feature or model serving. * Familiarity with the AWS ML ecosystem (S3, Glue, Redshift, SageMaker). * Experience with feature stores. * Experience with experiment tracking tools (MLflow, W&B). * Experience with GenAI and LLMs, such as speeding up data science work or building internal tools ## Description You'll be the first data scientist on the Analytics team, focused on non-risk problems. You'll be reporting to the VP, Analytics and will own the models behind customer acquisition, marketing efficiency, customer lifecycle, and operations, from feature engineering through validation, and contribute to our feature store on AWS. You'll work daily with analysts and with MLOps to deploy and run your models. MLOps owns deployment, infrastructure. You own how the models perform against business and statistical metrics. This is a high-visibility, hands-on role. You'll write production code and shape how Self does data science. What you'll do: * Build models that move the business. Design, build, and validate production-grade models for lead scoring, lifetime value, churn and retention propensity, next-best-action, message optimization, and time-series forecasting, working with Marketing, Product, and Operations leaders. * Take pilots to production and keep them working. Turn proven pilots into reproducible, production-ready code that MLOps can package, deploy, and schedule. Track experiments and code versions, and own validation, retraining, and refresh decisions using MLOps monitoring. * Engineer the features. Own feature engineering and contribute to our AWS feature store with MLOps. * Prove the impact. Measure and communicate the business impact of your work in dollars and percentages. * Set the technical bar. Define coding, validation, and operational standards for data science on the team. Favor simple, high-impact solutions over overengineered ones, and mentor analysts who want to move closer to data science