Senior Data Scientist

Position:  Senior Data Scientist Position Type: Full Time (Primarily Remote) Salary: $150,000-$165,000 DOE Join our team at Mainstay, a division of Lemnis! Lemnis is a public charity dedicated to harnessing transformative change to expand learning for all. We are excited to announce a new opportunity for a Senior Data Scientist to join our innovative, forward-thinking, and growing organization. At Lemnis, we are committed to fostering a collaborative and inclusive work environment where every team member can thrive. If you are passionate about expanding learning for all, eager to make a meaningful impact, and ready to take on new challenges, we would love to hear from you. Apply now and be a part of our journey! Position Summary  At Lemnis, we believe that the world is changing in exciting ways. It's up to us to create more equitable, flexible, and learner-centered systems that empower young people to rise to the challenges and opportunities of the future. Do you have questions? At Mainstay, our users have millions, so it’s imperative that our systems are stable, robust, and scalable. As a Senior Data Scientist at Mainstay, you will help implement features that help our partners and end users. About Mainstay  At Mainstay, we believe one conversation can spark a brighter future. Our Engagement Platform makes it easy for colleges and businesses to start and measure conversations that drive action at scale. From our rigorous research methods to our Behavioral Intelligence framework, everything we do is designed to help people take the next step toward achieving their goals. This is your chance to drive impact at Mainstay and for our users by building and enhancing the systems that power our product. We’ll provide the environment for you to master your skills and find both personal and professional growth. We are invested in promoting from within and providing the support and mentorship aimed at your long-term success. The engineering team is a group of talented full-stack developers, data engineers, analytics engineers, and product experts. Engineers collaborate closely with their counterparts in the product team and external stakeholders to build new functionality and enhance existing products to delight our partners and end users. The team is excited to continue tackling new challenges and leverage cutting-edge technologies to solve impactful problems. About the Role We're hiring a Senior Data Scientist to find the predictive signals in our data and turn them into insights our partners can act on. Your early work is less about squeezing out marginal accuracy and more about finding signals. You'll build student-level predictors that sync into partner systems, analyze conversational data at scale to make our AI smarter, improve access to reliable data and insights, and leverage internal tools to scale your expertise. As the senior data scientist on a lean team, you’ll have real say in what we build. We have more promising directions than capacity, so part of the job is deciding which signals are worth pursuing, which analyses will generalize, and which requests to decline. You’ll set the methodology bar for modeling and evaluation work here, and you’ll be the person others come to when they aren’t sure whether to trust a number. You'll work alongside our Senior Data Engineer and Senior Analytics Engineer. They own infrastructure, pipelines, and the shared semantic layer. You own predictive modeling, unstructured data analysis, AI evaluation, and the analytics surfaces that make data accessible and trustworthy. If you're looking to train models full-time, this isn't that role. What You'll Do Find and ship predictive signals Investigate signals across student engagement, outcomes, and partner health, prioritizing by business impact over technical interest. Build predictive models that reach partners through the systems they already use, where output drives real action. Set thresholds against the alert volume teams can actually act on, and be explicit about the cost of a false positive when predictions reach a partner. Evaluate performance across student populations, not just in aggregate, and document known limitations alongside the model. Monitor deployed models for drift, and retire models that stop earning their place. Know when not to build a model. Some questions are better answered with an analysis, a definition change, or a conversation. Own scheduling and monitoring for your models, using orchestration the whole team can maintain. Analyze our conversational and unstructured data Apply embeddings, clustering, and classification to conversational data, support and service records, and other unstructured sources to surface themes, gaps, and emerging concerns. Turn what you find into changes that improve the product and the partner experience. Build the text analysis foundations that make our unstructured data retrievable and useful to AI tooling. Own our AI data tooling and evaluation Own our in-warehouse AI configuration, including verified queries, prompts, and agent tooling, and the semantic views that expose your model output. You'll partner with an Analytics Engineer where this depends on the shared semantic layer. Build and maintain the AI evaluation framework for data team work: rubrics precise enough that two reviewers agree, a defensible sampling approach, and reporting that shows whether a model or prompt change actually improved anything. Set the evaluation standard for data team work, and partner with Product and Engineering to share evaluation methodology more broadly. Make your work reusable Equip internal teams with the data and analysis they need for our most strategic partners, favoring work that generalizes over one-off requests. Grow into building the tooling and training that lets those teams answer questions without the data team. Document reasoning, assumptions, and tradeoffs in our internal data knowledge base as part of finishing the work. Work in dbt/code alon