Human-in-the-Loop Data Scientist - Quality Analytics

Sperry Rail; Shelton (CT); On-site; USD 90,000 - 130,000; Full time

Job summary

Sperry Rail is seeking a data scientist to build retrospective scoring of analyst dispositions and quantify agreement across teams. You will work closely with risk analytics, operations, and engineering to understand how detection thresholds and data quality affect outcomes.

You will explain methods to non-technical stakeholders, manage complex analyses with imperfect labels, and contribute to tooling and training content. The role sits near operations and emphasizes actionable insights.

Qualifications

• Strong proficiency in Python and SQL with data science libraries.
• Applied statistics, including metrics, sampling, bias, and experimental design.
• Supervised ML on imbalanced, imperfectly labeled data.
• Experience evaluating decisions against ground truth that arrives late or is incomplete.
• Ability to explain method and uncertainty to an operational audience.
• Familiarity with Git, CI/CD, and agile development practices.
• Strong problem-solving and quick learning.
• Clear communication with non-technical stakeholders.

Responsibilities

• Build retrospective scoring of analyst dispositions against the data available at the time of review.
• Quantify agreement and variation across analysts, shifts, territories, and test conditions.
• Separate analyst-attributable outcomes from detection-threshold, data-quality, and volume effects.
• Identify contributing factors behind missed and misinterpreted defects and express them as a reusable taxonomy.
• Turn findings into feedback loops: training content, review guidance, tooling changes, and threshold recommendations.
• Define and maintain measures of decision quality and their limits.
• Design sampling and review studies where passive data cannot answer the question.
• Collaborate with detection and platform teams to ensure signals are captured at source.
• Present findings to analysis leadership and the wider engineering organization.
• Write clean, tested, well-documented code and participate in code reviews and sprint planning.
• Write and maintain documentation so the methodology is transferable.

Skills

Python

SQL

Statistics

Imbalanced data

Ground truth

Communication

Version control

Problem solving

Teamwork

Education

Bachelor's degree

Advanced degree welcome

Tools

Git

CI/CD

Agile

Job description

Sperry Rail is seeking a data scientist to build retrospective scoring of analyst dispositions and quantify agreement across teams. You will work closely with risk analytics, operations, and engineering to understand how detection thresholds and data quality affect outcomes.

You will explain methods to non-technical stakeholders, manage complex analyses with imperfect labels, and contribute to tooling and training content. The role sits near operations and emphasizes actionable insights.