About this role

Data Engineer / Document AI Engineer

Required qualifications

Five or more years building production data pipelines, with at least two years on document processing, OCR, or information extraction at scale.
Strong Python. Comfortable owning a pipeline from ingestion through storage without handoffs.
Hands-on experience with a cloud document intelligence service. Azure AI Document Intelligence preferred; AWS Textract or Google Document AI experience transfers.
Practical image preprocessing for degraded scans: deskew, denoise, binarization, contrast handling. You have made bad faxes readable before.
Structured extraction from unstructured text using large language models, including prompt design for reliable field-level output and handling of non-deterministic responses.
Entity resolution or record linkage experience. You understand why fuzzy name matching alone fails and what to do about it.
SQL and relational data modeling, including modeling typed relationships between records.
Experience measuring data quality quantitatively: error rates, confidence calibration, sampling design, and gold-set construction.
United States citizenship and eligibility for federal public trust suitability screening.

Preferred qualifications

Azure Government or Microsoft 365 GCC delivery experience, including working within FedRAMP-authorized service constraints and handling Controlled Unclassified Information.
Microsoft Graph API and SharePoint Online at scale.
Azure Functions, Azure SQL Database, Azure Blob Storage, Microsoft Fabric.
Power Automate for event-driven triggers.
Prior federal contract delivery, particularly against a fixed-price schedule with defined acceptance gates.
Knowledge graph construction or graph query experience.

Data Engineer || United States citizenship is required

Key skills