Senior Data Scientist

About Omnidian Omnidian, Inc. is a fast-growing Series C tech-enabled service company revolutionizing performance assurance for the distributed solar and energy storage industries. Omnidian is building a more sustainable future for the planet through our passionate teams, our innovative technology, and by creating an amazing customer experience. We are a certified B Corp, headquartered in Seattle, WAThe Job We are seeking a Senior Data Scientist to design, build, and scale advanced detection and diagnostic models for solar and storage assets. In direct support of Omnidian’s mission, your work will enable faster detection, fewer truck rolls, and better customer experience. As a key member of the data science team, you will report to the Director of Data Science and partner closely across product management, engineering, and operations to take models from idea to production in Omnidian’s Resolv platform.\nWhat You'll DoAt Omnidian we believe in trust and autonomy. How you create an impact is ultimately up to you. Here is an outline of some of the things you’ll be doing: Lead Issue Detection & Diagnostic Models (60%) Develop machine learning models that detect underperformance, faults, and anomalies in PV and storage system time-series data Build diagnostic logic that distinguishes root causes (soiling, shading, inverter clipping, string outages, communication gaps, sensor drift, degradation) of underperformance. Own the data science lifecycle from problem definition and model development through validation and monitoring, partnering with Engineering to deploy models into production. Help shape the technical direction for the detection and diagnostics roadmap and strengthen the team’s approach to model design, validation, and testing Partner to Deploy High-Impact Models (20%) Partner with engineering to integrate your models into Resolv, turning model outputs into concrete, prioritized dispatch recommendations Close the loop with operations: track how your recommendations perform in the field — truck rolls avoided, issues resolved faster — and use that data to improve your models Ideate and Explore New Models (20%) Explore and prototype new modeling approaches and validate their business impact before investing in production deployment Stay current with advances in ML and energy analytics, and share what you learn with the team Who You Are You own outcomes and measure success by the decisions your models improve You are rigorous with messy, real-world data. You dig into the quality, provenance, and business context of the data your models consume You take your models through the full lifecycle (from problem framing, prototyping, and validation to deployment and monitoring) You translate across domains. You work fluently with engineers, operations teams, and customers to scope what data science can (and can't) solve, and you can explain a diagnostic model to someone who has never heard of state-space methods. Experience You’ll Need MS in a quantitative field or equivalent experience; 6+ years in applied data science Strong hands-on experience with time-series methods: anomaly detection, forecasting, seasonal decomposition, state-space models Production ML experience including model deployment, monitoring, and retraining pipelines Proficiency in Python (pandas, numpy, scikit-learn) and SQL; experience with at least one of PyTorch/TensorFlow Experience working with noisy, irregularly sampled, multi-sensor data at scale Track record of leading modeling projects end-to-end — translating ambiguous business problems into measurable model outputs and shipping them to production Proven written and verbal communication skills The ability to work with large cross-functional team. Experience mentoring data scientists or engineers and raising team standards Experience with architecture and tooling decisions for production ML systems on a major cloud platform (e.g., AWS, Azure, GC) Experience That’s a Plus GenAI: Experience developing with Claude Code or putting LLMs into operational workflows Solar Modeling: Familiarity with PV performance modeling tools (e.g., pvlib, PVsyst) or irradiance and weather datasets. Hands-on work with inverter, SCADA, or other industrial IoT data streams Tooling: Experience with Databricks, MLflow, or similar ML platform and orchestration tooling Forecasting: Experience with probabilistic forecasting or failure prediction — e.g., forecasting energy production, predicting part failure, or estimating remaining life Optimization: Experience with ranking, prioritization, or decision-optimization problems such as alert triage or field-dispatch scheduling Logistics We plan to have this role start in early Fall 2026 We are unable to provide sponsorship for this role, now or in the future Most of our roles offer the opportunity to work remotely If you are in the Seattle area, we offer a vibrant office space in the historic and beautiful Smith Tower, in the heart of Pioneer Square We prioritize app