Applied Data Scientist

This is a fully remote position, open to applicants in United States.

📋 Description

• Develop and sustain models that assess incoming requests in real-time — evaluating quality, identifying risk indicators, and directing items to the appropriate handling process prior to human review.

• Train classification and ranking models using historical outcome data (accepted, declined, loss events) to forecast account quality and prioritize review queues — akin to fraud scoring in fintech, patient risk stratification in healthcare, or lead scoring in high-traffic sales platforms.

• Incorporate structured and unstructured third-party data signals as features in models: geospatial layers, firmographic information, external risk indicators, and fields extracted from documents.

• Deliver model outputs through API to ensure scores and flags are integrated seamlessly within the workflow tools utilized by the operations team — positioning your model as a product feature rather than merely a report.

• Convert business rules and decision-making criteria into machine-executable logic applicable programmatically at intake — transitioning decisions currently requiring human judgment into automated or assisted processes.

• Construct and manage feature engineering pipelines that support these models: normalizing inputs, addressing missing data, encoding categorical variables, and enhancing records with external data sources.

• Create model explainability layers to ensure end users comprehend the reasoning behind a record's scoring or routing — essential for user trust and regulatory compliance in our industry.

• Oversee the entire deployment lifecycle: containerizing models, developing inference APIs, collaborating with engineering for production integration, and establishing monitoring for model drift and performance degradation over time.

• Build pipelines for extracting structured data from unstructured documents: forms, PDFs, emails, and attachments that are part of the intake workflow. Utilize NLP and LLM-based extraction techniques to minimize manual data entry and enhance the completeness of records entering the decision workflow.

⛳️ Requirements

• 3+ years of experience as a data scientist or ML engineer, with multiple production deployments where your model was utilized by actual end users.

• Strong Python proficiency with production-quality coding practices: modular, tested, version-controlled code — not limited to notebook-quality work.

• Practical experience with ML frameworks (scikit-learn, XGBoost, LightGBM, PyTorch, or TensorFlow) and applied understanding of classification, ranking, regression, and feature engineering for real-world, noisy datasets.

• Experience in building and maintaining data pipelines that supply production models — scheduled, monitored, and reliable, rather than just ad hoc EDA scripts.

• Familiarity with model deployment methodologies: REST APIs (FastAPI or Flask), containerization (Docker), and cloud deployment on AWS, GCP, or Azure.

• Proficient in SQL; capable of extracting and transforming data from a cloud warehouse (Snowflake, BigQuery, or Redshift) as part of feature engineering workflows.

• Strong problem-framing abilities: you can take an ambiguous business problem, determine if ML is the appropriate tool, define the target variable, and outline the modeling approach before writing any code.

🏝️ Benefits

• A collaborative, results-driven environment

• Competitive compensation and comprehensive benefits

• Year-round social and community events

• Ongoing mentorship and professional development

• Endless opportunities for upward mobility