Scope of Work – Project-Based Data Engineer (LATAM Preferred)
Project Overview
Forward Storage is seeking a contract-based Data Engineer to design, implement, and document a modern data warehouse and ETL/ELT architecture. The goal is to centralize operational, financial, sales, and marketing data into an analytics-ready warehouse to support Tableau/Power BI reporting.
This engagement is project-based, with a clearly defined build phase followed by optional light ongoing support.
Primary Objectives:
• Design and implement a scalable, low-maintenance data warehouse.
• Establish automated data pipelines from core SaaS platforms.
• Model data into analytics-ready fact and dimension tables.
• Ensure data accuracy, reliability, and documentation for long-term ownership by the internal analyst.
Initial Data Sources:
• Cubby – property management (financial and operational data).
• AppFolio – property management (financial and operational data).
• HubSpot – CRM, leads, sales funnel data.
• Google Ads – campaign, spend, performance data.
• Facebook Ads – campaign, spend, performance data.
Preferred Technology Stack (Open to Vetting):
• Data Warehouse: Google BigQuery (preferred), Snowflake or equivalent.
• ELT / Ingestion: Airbyte (preferred), Fivetran, Stitch, or equivalent.
• Transformation Layer: dbt (Core or Cloud preferred).
• BI Tool: Tableau (preferred), others to be considered.
• Version Control: GitHub or GitLab.
Note: The engineer may recommend alternative tools if they better meet reliability, cost, or maintainability goals. Final stack selection will be mutually agreed upon.
Scope of Work
Phase 1 – Discovery & Architecture (1–2 weeks)
• Review available APIs, data schemas, and access methods for all source systems.
• Recommend final warehouse and ELT architecture.
• Define data ingestion strategy (incremental loads, refresh cadence).
• Establish naming conventions, schemas, and data modeling standards.
• Define high-level data governance and quality approach.
Phase 2 – Implementation & Modeling (3–5 weeks)
• Configure cloud data warehouse environment.
• Build automated ELT pipelines for all Phase 1 data sources.
• Create raw/staging tables with minimal transformation.
• Develop transformed models including:
* Financial metrics by property and time.
* Operational performance (occupancy, units, activity).
* Sales and funnel metrics from HubSpot.
* Marketing spend and performance by channel.
• Design analytics-ready fact and dimension tables.
• Implement incremental refresh logic and basic data validation tests.
Phase 3 – QA, Documentation & Handoff (1–2 weeks)
• Validate data accuracy with the internal analyst and stakeholders.
• Optimize queries and model performance.
• Deliver documentation including:
* Data dictionary.
* Entity-relationship overview.
* Pipeline refresh schedule.
• Walkthrough and handoff to internal analyst.
• Finalize Git repository and project artifacts.
Out of Scope
• Advanced ML or predictive modeling.
• Real-time streaming architecture (unless separately agreed).
• Ongoing dashboard development.
• Long-term infrastructure monitoring beyond agreed retainer.
Deliverables
• Production-ready data warehouse.
• Automated ELT pipelines.
• Analytics-ready data models.
• Documentation and handoff materials.
• Optional support transition plan.
Timeline
• Estimated total duration: 6–8 weeks.
Compensation & Engagement Model
• Fixed project budget: USD TBD.
• Milestone-based payments preferred.
• Optional ongoing support retainer: 5–10 hrs/month.
Required Qualifications
• 5+ years of data engineering or analytics engineering experience.
• Strong SQL and data modeling skills.
• Experience with cloud data warehouses (BigQuery, Snowflake, etc.).
• Experience with ELT tools (Airbyte, Fivetran, dbt, etc.).
• Familiarity with SaaS data sources (CRM, Ads platforms, financial systems).
• Comfortable working independently in a remote environment.
• Clear written and spoken English.
Success Criteria
• Reliable, automated data refreshes.
• Clean, documented, analytics-ready data models.
• Minimal ongoing engineering dependency.
• Smooth handoff to internal analyst.
