Data Architect: Construction Catalog Cleaning, Cost Code Mapping & Postgres Schema Design

Job Description: We are looking for an experienced Data Architect / Lead Data Analyst to standardize, restructure, and calculate financial metrics for a large dataset of 150,000+ construction catalog items. In this role, you will clean raw catalog data, research item specifications, disaggregate bundled items into distinct labor and material rows, update unit costs and margins, assign a QuickBooks-compatible Cost Code Schema, and design an optimized PostgreSQL database schema. Note: You will not need direct access to our live database, nor do you need to write custom ingestion code. Our internal dev team will handle the actual data importing. Your focus is delivering the cleaned/restructured data, the database architecture (DDL), and migration files. Key Responsibilities: Data Cleaning & Standardization: Process, deduplicate, and normalize 150,000+ catalog rows of construction materials, equipment, and labor line items. Labor & Material Disaggregation: Identify bundled assembly items and split them into separate, standalone rows for Labor and Materials with correct classifications. Cost & Margin Updates: Research, assign, and update accurate base unit costs, default margins (percentages/markups), and final sell prices for every individual row. Cost Code Mapping & Research: Mine item descriptions to accurately assign a standardized Cost Code hierarchy that mirrors our accounting structure in QuickBooks. Database Schema Design: Design a clean, scalable, relational schema for PostgreSQL to store catalog items, item types (labor vs. material), pricing/margin rules, cost codes, and metadata. Migration File Creation: Provide well-structured SQL migration files/DDL scripts (.sql) for table creation, constraints, foreign keys, data types, and indexes. Key Deliverables: Cleaned & Restructured Dataset: Standardized file (CSV/Parquet) containing the catalog items split into individual labor and material rows, updated with unit costs, margin logic, and mapped QuickBooks cost codes. Postgres DDL / Schema Files: .sql files defining the PostgreSQL table structures, relationships (e.g., items, cost codes, pricing rules), data types, and indexing strategy. Migration Scripts: Clean SQL migration files for any schema updates, patches, or schema versioning. Required Qualifications & Skills: Proven Data Architecture / Modeling Experience: Deep understanding of relational database design, data normalization, and schema definition in PostgreSQL. High-Volume Data Manipulation: Advanced proficiency using Python (Pandas/Polars), SQL, or DuckDB to manipulate and restructure 150k+ row datasets efficiently without data corruption. Construction & Cost Estimating Knowledge: Familiarity with construction industry terminology, labor vs. material breakdowns, and standard cost coding frameworks (e.g., CSI MasterFormat, NAHB, etc.). Financial Data Logic: Experience working with unit costs, margins, markups, and mapping data to QuickBooks accounting/cost code structures. How to Apply: Please submit your proposal along with: A brief summary of a similar large-scale catalog cleaning, pricing restructuring, or data modeling project you’ve completed. The primary tools/tech stack you plan to use to process, split, and clean the 150,000+ rows. To confirm you’ve read the full description, please start your cover letter with the word: "BUILD".