This repository documents my six-month AI certification and enterprise AI-agent development project.
Design and build a secure enterprise AI assistant capable of finding organizational information, explaining answers with citations, and eventually performing explicitly authorized actions.
This project accompanies preparation for:
- Microsoft Azure AI Fundamentals
- Microsoft Applied Skills for AI agents
- PMI Certified Professional in Managing AI
- Microsoft Azure AI Apps and Agents Developer Associate
Completed:
- Introduction to AI concepts
- Introduction to generative AI and agents
Completed:
- Introduction to natural language processing concepts
- Introduction to AI speech concepts
Completed:
- Introduction to computer vision concepts
- Introduction to AI-powered information extraction concepts
- Introduction to retrieval-augmented generation concepts
The portfolio contains the architecture, project-management documentation, hands-on exercises, and knowledge sources developed throughout the certification curriculum.
The initial knowledge base contains six fictional enterprise documents representing policies, procedures, and technical guidelines.
These documents are used to demonstrate how enterprise information can be prepared, retrieved, and used to support grounded AI responses.
Week 3 — Vision, Information Extraction, and RAG
Current work includes:
- Documenting the initial RAG architecture.
- Creating a synthetic enterprise knowledge base.
- Demonstrating manual retrieval using enterprise documentation.
- Constructing grounded responses supported by source citations.
The initial RAG design establishes the conceptual process for document ingestion, chunking, indexing, retrieval, and answer generation.
Automated retrieval, AI model integration, and production AI services have not yet been implemented.