Staff Data Scientist

HR Acuity LLC•Remote,

2d•$170,000 - $190,000•Remote

About The Position

We are seeking a Staff Data Scientist who is deeply curious about our product, our customers, and the employee relations data that sits at the heart of what we do. Reporting to the Director of Platform and Data Engineering, you will own the development of an authoritative, end-to-end understanding of our data: what it represents, how it’s structured, where it’s reliable, and what it can tell us about workplace issues at scale. From that foundation, you’ll build the analytical models and customer-facing data products that help organizations understand patterns in their employee relations (ER) programs and benchmark against industry peers. This is a high-impact role where genuine product curiosity isn’t a nice-to-have; it’s the prerequisite for doing the work well. You'll bring rigorous statistical thinking and strong engineering fundamentals to everything from building the models that power our product to presenting benchmark findings directly to customers.

Requirements

Bachelor's or Master's degree in Statistics, Data Science, Computer Science, or a related quantitative field — or equivalent practical experience
6+ years of experience in data science, applied statistics, or machine learning
Proven track record building and deploying predictive models in production environments
Comfortable establishing model lifecycle practices in greenfield environments, including experiment tracking, versioning, and drift monitoring
Strong Python skills across the data science ecosystem (pandas, scikit-learn, statsmodels, scipy, etc.)
Solid SQL command with experience working with relational databases at scale
Experience with Snowflake and cloud-based ELT patterns
Azure experience preferred, particularly for model pipelines and container-based workloads (e.g., Kubernetes)
Demonstrated experience with unstructured text data, including NER, PII detection and anonymization pipelines, text classification, topic modeling, entity extraction, and embeddings
Experience building or maintaining systems that consistently replace identifying information across large document corpora is strongly preferred
Experience evaluating LLM-powered features in production, including assessing AI-generated outputs for accuracy, calibration, and bias
Familiarity with prompt review, golden dataset design, or model quality monitoring is strongly preferred
Proficiency in experiment design and A/B testing methodology, including power analysis, significance testing, and translating results into product decisions
Experience building and communicating data-driven benchmarks, scorecards, or comparative analytics, ideally in a B2B or SaaS context
Experience working with sensitive or compliance-adjacent data (HR, healthcare, legal, or financial) is strongly preferred
Familiarity with privacy-preserving statistical methods for aggregate analytics
Familiarity with statistical disclosure control techniques such as minimum cohort thresholds, suppression rules, and aggregation standards for sensitive data
Demonstrated ability to translate complex statistical concepts into clear, actionable narratives for varied audiences, in both written and visual formats
Familiarity with data visualization tools such as Tableau, Power BI, Plotly, or Altair

Nice To Haves

Familiarity with dbt or similar transformation frameworks is a plus

Responsibilities

Develop and continuously deepen a comprehensive understanding of HR Acuity's full data landscape, including case types, taxonomies, workflows, and customer segments, grounded in the real-world employee relations problems the data represents
Assess what makes HR Acuity's dataset competitively distinctive, identify gaps or additions that would strengthen its analytical value, and proactively surface opportunities that extend beyond the current product roadmap
Build rigorous documentation, lineage, and data quality baselines for HR Acuity's ER dataset, establishing the foundation the analytics product needs to grow on
Design and build statistical models and machine learning solutions that surface trends, risk signals, and benchmarks from HR Acuity’s employee relations case data, with rigorous attention to bias detection and mitigation given the sensitivity of the underlying data
Apply NLP and text analytics techniques to extract structure, patterns, and signals from HR Acuity’s unstructured case data, including case notes, investigation narratives, and incident descriptions; this includes building and improving anonymization pipelines for sensitive unstructured content, with a focus on preserving analytical value while protecting identity
Own and elevate HR Acuity’s customer-facing analytics and benchmarking capabilities: deepen their statistical rigor, expand the insights they deliver, and ensure they reflect the full analytical depth of the underlying dataset; this is the current core product, and making it more statistically meaningful and defensible is a primary responsibility of this role
Design and execute experiments to measure the impact of product and analytical changes, applying rigorous statistical methods, including power analysis and significance testing, to feature evaluation and continuous improvement
Collaborate with Data Engineering to ensure pipelines, schemas, and data models support analytical needs and maintain data quality
Partner closely with Product and Engineering to define metrics, evaluate feature performance, and translate analytical findings into product decisions
Translate complex analytical findings into clear, compelling narratives for non-technical stakeholders, including customers and internal leadership
Apply responsible data practices throughout all work, prioritizing privacy, fairness, and ethical handling of sensitive employee information
Define and maintain the statistical evaluation framework for AI and ML features, including golden datasets, quality metrics, and release gates for data quality
Contribute to technical approach decisions for AI features, bringing data-driven judgment on when classical ML, statistical models, or rules-based solutions are more appropriate than LLM-based approaches
Serve as the statistical and data quality voice on LLM-powered features, evaluating whether AI-generated insights, summaries, and recommendations are statistically sound, appropriately calibrated, and meaningful for customers
Review prompt designs and model outputs for accuracy, bias, and statistical validity, partnering with engineers to iterate before and after release
Monitor model and data quality over time, identifying and addressing degradation, drift, or unexpected changes in underlying distributions
Establish data science standards, documentation practices, and reproducibility norms that scale as the team grows
As the data science function grows, provide mentorship and technical guidance to junior practitioners, contributing to a culture of rigor and continuous learning