Principal Data Scientist

Risepoint is building an AI-powered platform to improve student outcomes across a network of universities. As a Principal Data Scientist, you will lead end-to-end AI/ML initiatives, data engineering, and production delivery to build predictive models and the Next Best Experience platform, driving retention, engagement, and enrollment across more than 100 university partners. This is a remote role.

Responsibilities

• Set direction for and lead AI/ML initiatives end-to-end—scoping ambiguous business opportunities, defining the problem and success criteria, designing the technical approach, managing implementation, and driving outcomes—coordinating across Product, Engineering, CX, Partnership, and university partner teams.
• Own accountability for delivering measurable business outcomes from each initiative: retention lift, engagement improvement, enrollment conversion, and pipeline efficiency.
• Drive alignment and decision-making across teams at each stage of an initiative’s lifecycle, resolving moderately complex, cross-functional problems independently and proactively while escalating only when tradeoffs require leadership decision.
• Identify and scope net-new AI/ML opportunities that deliver impact for students, university partners, and Risepoint’s business; frame options, recommend a path forward, and advocate for prioritization with leadership.
• Manage relationships with key vendors and software providers as a workstream leader, ensuring delivery commitments are met.
• Influence peers, managers, and senior stakeholders across BT and adjacent business functions—including Partnership and Customer Experience—by translating technical tradeoffs into business implications and building support for shared decisions without direct authority.
• Build and deploy predictive models—including churn risk, engagement propensity, and success likelihood—that power proactive student outreach and are monitored continuously in production.
• Lead the design and implementation of “next best action” logic in close partnership with Product and CX, from logic design through production deployment.
• Prototype, test, and productionize models using MLOps frameworks (Databricks, MLFlow, dbt, Dagster), owning the full model lifecycle.
• Own clean, reliable data pipelines and feature stores that support model development and production deployment at scale, doubling as the data engineer for the workstream.
• Work with speech analytics and structured CRM/LMS data to derive behavioral insights across the student lifecycle.
• Architect, build, and own scalable, reliable data pipelines and the underlying data infrastructure (lakehouse, warehouse, and feature stores) end-to-end—operating as the team's principal data engineer.
• Design and maintain data models, ELT/ETL workflows, and feature pipelines that serve both analytics and production model-serving needs.
• Take models to production and keep them healthy there: own packaging, deployment, serving, versioning, and the full production lifecycle, including rollback.
• Automate production workflows with orchestration tools (Dagster, Airflow) for scheduling, dependency management, and pipeline reliability.
• Implement CI/CD pipelines and infrastructure-as-code (Terraform, Docker, Kubernetes) to automate testing, deployment, and reproducible environments.
• Build automated monitoring and observability—data-quality checks, model and data drift detection, alerting, and automated retraining triggers—to keep production systems running with minimal manual intervention.
• Own data quality, governance, lineage, and cost/performance optimization across the platform, setting the engineering standards the team builds against.
• Design and lead A/B testing programs to measure model-driven impact on retention, engagement, and satisfaction, owning the decision to ship, iterate, or stop.
• Establish feedback loops and real-world performance monitoring frameworks that enable continuous model improvement.
• Translate complex technical findings into clear, executive-ready narratives that drive cross-functional alignment and action.
• Mentor data scientists and engineers across the team and raise the organization’s technical bar through code reviews, pair work, and knowledge-sharing.
• Model ownership, adaptability, and technical leadership in a fast-changing environment; set the standard for what it means to own a domain end-to-end.
• Define technical approaches and promote technical best practices across teams, including standards for data lineage, traceability, and explainability that support user trust and regulatory needs.
• Champion a continuous-learning environment, driving adoption of and experimentation with the latest AI-assisted coding and collaboration tools to multiply team velocity.
• Influence the data science and AI roadmap as the technical expert and thought leader to Product and Engineering leadership.
• Predictive models are deployed, monitored, and demonstrably improving student outcomes (e.g., reduced churn, higher engagement rates)—and you can point to specific initiative decisions you made that drove those results.
• Cross-functional partners describe you as a principal-level technical leader who owns outcomes, not just analysis—independently resolves moderately complex problems, builds alignment across functions, manages implementation, and delivers results.
• Experiment programs are well-designed, velocity is high, and a clear percentage of tests yield statistically significant outcomes that inform production decisions.
• The data foundation is materially stronger because of your workstream ownership: pipelines are cleaner, features are better documented, and the team ships faster.
• You actively raise the organization’s technical standard, establishing best practices others adopt, and mentoring data scientists and engineers toward greater ownership and impact.

Requirements

• A proven track record of delivering measurable consumer and business impact through AI/ML initiatives—scoping, managing implementation, and owning outcomes end-to-end.
• Experience operating as a principal-level technical leader or domain authority: independently resolving moderately complex, ambiguous problems; setting direction for AI/ML programs; and delivering outcomes across teams in a cross-functional environment.
• 8+ years in applied machine learning or data science, ideally in education, consumer tech, personalization, or a complex behavioral domain.
• Strong background in predictive analytics, recommendation systems, and experimentation (A/B testing, causal inference, uplift modeling).
• Deep expertise in Python and SQL; proficiency with ML libraries (scikit-learn, XGBoost, TensorFlow, or PyTorch).
• Experience with Databricks, MLFlow, dbt, and Dagster—or demonstrated ability to ramp quickly on a modern MLOps stack.
• Principal-level data engineering experience: architecting and operating production data pipelines, data models, and feature stores at scale.
• Hands-on experience taking models to production and operating them there—deployment, serving, monitoring, and retraining.
• Proficiency with production automation tooling: workflow orchestration (Dagster, Airflow), CI/CD, infrastructure-as-code (Terraform), and containerization (Docker, Kubernetes).
• Strong grounding in data quality, governance, lineage, and observability practices.
• Comfort working with complex, multi-source datasets (CRM, LMS, communication logs, speech analytics).
• Excellent communicator and influencer across technical and non-technical audiences, including peers, managers, executives, and business partners outside BT; you make the science accessible without losing rigor and build support for decisions through evidence, clarity, and trust.
• Bachelor’s or Master’s degree in a technical discipline (computer science, statistics, econometrics, mathematics, or engineering).

Technologies

• Python
• SQL
• scikit-learn
• XGBoost
• TensorFlow
• PyTorch
• Databricks
• MLFlow
• dbt
• Dagster
• Airflow
• Terraform
• Docker
• Kubernetes

How Impact Will Be Measured

• Business outcomes tied to model-driven initiatives: retention rates, re-engagement rates, enrollment completion, and conversion lift.
• Initiative delivery: on-time scoping, cross-functional execution, and outcome realization against defined success metrics.
• Model performance metrics: accuracy, precision, recall, and AUC across deployed models; degradation alerts and retraining cadence.
• Production reliability and automation: pipeline uptime, data-quality SLAs, deployment frequency, and reduction in manual intervention.
• Experiment velocity and signal rate: number of A/B tests shipped per quarter and percentage yielding statistically significant, actionable results.
• Qualitative feedback from Product, Engineering, CX, Partnership, and Customer Experience partners on initiative ownership, communication quality, cross-functional influence, and effectiveness in resolving moderately complex problems independently.