Azure AI Engineer - Document Intelligence & Enterprise Search

• Configuring Azure OpenAI so that data retention and abuse-monitoring review are disabled and no customer data leaves the authorization boundary
• Prior federal or state government delivery; Section 508 accessibility
• Microsoft certifications: AI-102, AZ-305, DP-600

What this role is not

Not a data-science or model-training role. You will not train, fine-tune or retrain any model on customer data - that’s contractually prohibited. This is applied engineering with Azure’s managed AI services inside a locked-down government tenant.

Pay: $70.00 - $90.00 per hour

Application Question(s):

• Experience using Azure Government or GCC
• Azure AI Document Intelligence
• Azure AI Search
• Azure OpenAI Service
• What’s your Work Authorization
• Total number of years in USA

• Hands-on production delivery with Azure AI Document Intelligence (formerly Form Recognizer) - custom extraction models, confidence scoring, OCR quality management on real scanned material
• Hands-on production delivery with Azure AI Search - indexing pipelines, skillsets, semantic ranking, vector search
• Azure OpenAI Service - embeddings and retrieval-augmented generation, including grounding responses in a document corpus with citations
• Document classification at corpus scale, and honest experience with what OCR accuracy actually looks like on degraded scans
• SQL and Python
• Able to work entirely remotely on US business hours and eligible to undergo a Tier 1 / Public Trust background investigation

Strongly preferred

• Work inside Azure Government / M365 GCC or GCC High - not just commercial Azure
• Microsoft Purview for data governance and classification

$70 - $90 an hour - Contract

A civilian federal agency is converting roughly 16,500 scanned PDF documents (~49,500 pages) of historical certification records held in SharePoint into a structured, searchable knowledge repository. You’d own the AI engineering across both phases