Job Description & Details
This is a hands-on contractor role building enterprise-grade AI solutions specifically within the Microsoft and Azure ecosystem. If you are tired of writing toy LLM wrappers and want to wire up actual RAG architectures and autonomous agent workflows for a business, this gig is worth looking into. You will be knee-deep in Azure, Copilot Studio, and Databricks, making legacy enterprise data actually searchable and useful through modern AI.
What You'll Actually Be Doing
Expect to spend your days architecting integrations between OpenAI endpoints, enterprise data sources, and Microsoft 365 environments like SharePoint and Teams. You will build and scale RAG pipelines, write robust Python services to manage API connections, and implement agent-based automations using Model Context Protocol (MCP) servers. The real challenge here isn't just writing prompts—it is dealing with messy enterprise data, ensuring secure API interactions, and turning disparate business systems into cohesive, intelligent workflows.
The Core Tech Stack
You are going to need rock-solid Python skills to build backend services and handle API integrations via REST, webhooks, and JSON. On the AI side, deep familiarity with Azure OpenAI, Databricks (including Databricks Genie), and prompt engineering is non-negotiable. Because this sits squarely in the Microsoft stack, you also need proven chops with Copilot Studio, Power Platform, and M365 infrastructure to connect the dots between models and users.
Interview Expectations
Expect the engineering team to test your API design and system integration capabilities under the hood of an AI application. They will likely ask you to whiteboard how you would design a secure, low-latency RAG pipeline that pulls data from a heavily permissioned SharePoint instance without leaking sensitive records. The interviewer is secretly looking to see if you understand the actual failure modes of LLMs in production, such as handling token limits, managing context windows, and dealing with hallucinations when enterprise data changes unexpectedly.
Application Advice
Your resume needs to heavily feature concrete examples of production-ready AI applications, not just school projects or tutorials. Make sure keywords like Azure OpenAI, RAG, Python, Databricks, and Copilot Studio are explicitly called out in your experience bullets to pass the automated filters. Don't just list technologies; write brief sentences showing how you used them to solve real infrastructure or workflow problems.