The Role

We are looking for a Data Scientist to join our Applied AI team. This is a hands-on, generalist role working across Arcadia’s core ML and AI workstreams: utility data extraction, consumption forecasting, and agent-powered production workflows. You will work directly with senior data scientists and engineering to build, evaluate, and ship models that drive meaningful impact on central business operations.

This role is a strong fit for someone early in their career who wants broad exposure to applied ML—from building classifiers and forecasting models to tuning prompts and supporting evaluation frameworks.

What You’ll Do

• Develop and extend production ML models for utility bill data extraction, including classifier and routing logic, LLM extraction prompts, and template confidence and accuracy analysis
• Build and iterate on forecasting models for operational use cases: predicting bill availability, estimating utility spend, and surfacing prioritization signals
• Contribute to audit and anomaly detection work—analyzing flagged bills, quantifying error rates, and helping tune tolerance thresholds
• Participate in offline experimentation and champion/challenger evaluations; document results clearly and compellingly to guide business decisions
• Work with data teams and subject matter experts to source, understand, and validate data; surface data quality issues proactively
• Write clean, well-documented Python code; contribute to shared notebooks and model development pipelines
• Collaborate with engineering to support model integration and monitoring; contribute to design docs and relevant code reviews

What We're Not Looking For

This is not a research role, a pure NLP or LLM engineering role, or a data analytics/BI role. Candidates whose experience is primarily in dashboarding, reporting, or model-free data work are not a fit. Candidates who have only worked in academic or research settings without production deployment experience should be screened carefully.

What Will Help You Succeed

• 1–3 years of experience building and evaluating ML models in a professional or research setting
• Solid foundation in statistics, supervised learning, classification, regression, and evaluation methodology (precision/recall, cross-validation, holdout testing)
• Proficiency in Python (pandas, scikit-learn, numpy) and SQL
• Experience with time series data or forecasting is a plus
• Exposure to generative AI tools and LLM-based workflows
• Ability to work iteratively and independently—formulate assumptions, document them, and advance projects without constant direction
• Clear communicator: able to explain what a model does and what its results mean to both technical and non-technical stakeholders
• Bachelor’s or Master’s degree in Computer Science, Statistics, Mathematics, or a related field—or equivalent practical experience