Data Science & Business Intelligence Analyst

BlueTeam•Boca Raton, FL

2d•Onsite

About The Position

The Data Science & Business Intelligence Analyst serves as the company's primary quantitative resource, transforming data from project, financial, and customer systems into decisions leadership can act on. The role spans the full analytical range: data acquisition and modeling, recurring and ad hoc reporting, statistical and predictive modeling, and the applied use of artificial intelligence to extract structured information from the document-heavy workflows that drive this business. The analyst will work with structured and unstructured data, build and maintain the reporting layer, develop forecasting and predictive models, validate those models against actual results, and automate manual processes. The standard for this role is defensibility: every number produced must reconcile to its source system, and every model must be documented, tested against data it was not trained on, and explainable to a non-technical audience. Selecting the simplest method that answers the question is preferred over sophistication for its own sake. This position supports the executive team crossing all departments of the Company. Initial priorities will center on sales reporting, expanding to enterprise analytics as the reporting foundation matures.

Requirements

5 years of progressive experience in data analytics, data science, business intelligence, or a related quantitative field, including hands-on ownership of both reporting and predictive modeling work.
Strong analytical and problem-solving skills with experience interpreting large datasets.
SQL proficiency sufficient to write and optimize multi-table joins, aggregations, and window functions against a production database without assistance.
Working proficiency in Python or R for data manipulation, statistical analysis, and modeling (for example pandas, scikit-learn, stats models, or equivalent libraries).
Demonstrated experience building, validating, and putting into use at least one forecasting or predictive model that informed an operating decision.
Expert proficiency in Excel, including advanced formulas, pivot tables, dynamic financial and operational models, and AI‑assisted model development.
Proficiency in CRM platforms, with hands‑on experience across multiple systems; ability to navigate, maintain data integrity, and extract insights across different systems environments.
Proficiency in Power BI, including dashboard design, DAX formulas, and data modeling.
Working proficiency with current AI tooling applied to real analytical work, including prompt design, structured output, and validation of AI-generated results before use.
Excellent communication skills with the ability to translate data into clear insights.

Nice To Haves

Background in construction, restoration, insurance, or another project-based industry is not required but is a plus.
Experience building automated dashboards and reporting systems.
Demonstrated ability to use AI for prospect and client research, including synthesizing information from multiple sources into actionable sales intelligence and executive‑ready presentations.
Experience with cloud data platforms and pipeline orchestration
Experience with large language model APIs, retrieval methods, embeddings, and evaluation techniques.

Responsibilities

Collect, clean, validate, and transform data from multiple source systems, including project management, accounting, CRM, and field data collection platforms.
Develop dashboards, reports, and visualizations that provide actionable insights.
Analyze historical trends, operational performance, productivity metrics, and business outcomes, including job-level margin, estimate versus actual variance, backlog and pipeline conversion, win rates by client and business unit, and receivable aging and collection cycle time.
Create recurring and ad hoc reporting for leadership teams.
Identify patterns, risks, and opportunities through quantitative analysis.
Build and maintain the queries, extracts, and pipelines that feed the reporting layer, including API-based extraction from source systems.
Reconcile reporting to the general ledger and to source systems so that analytical output and financial reporting do not diverge.
Build forecasting models for revenue, backlog conversion, labor and equipment demand, and cash flow, accounting for the seasonality and catastrophe-driven volatility inherent to storm restoration work.
Design and interpret experiments, quantify statistical significance, and measure realized business impact against forecast.
Present quantitative findings with stated confidence, known limitations, and the reasoning behind method selection.
Apply large language model tooling to production analytical workflows, including structured data extraction from unstructured documents, classification, and summarization, rather than ad hoc manual prompting alone.
Develop AI-assisted processes to streamline reporting, research, document review, and decision support, with human review controls at each output stage.
Evaluate emerging AI capabilities and recommend practical business applications, including build versus buy assessment and cost per unit of output.
Build automated workflows that reduce manual effort and increase efficiency.
Ensure data accuracy, integrity, and consistency across reporting systems.
Support data governance initiatives, validation processes, and data quality improvements.
Partner with stakeholders to establish reporting standards and best practices.
Document data sources, transformations, model logic, and code so that all work is reproducible by someone other than the author.
Partner with business leaders to understand strategic priorities and deliver data-driven recommendations.
Present findings to both technical and non-technical audiences.
Support initiatives across Sales, Operations, Finance, Marketing, and Executive Leadership.
Translate complex analyses into actionable business recommendations.
Challenge analytically unsupported conclusions, including those already held by leadership, and state plainly where available data is insufficient to answer the question asked.