Senior AI Engineer / Architect @ Microsoft
Designing agentic, event-driven AI systems that integrate tools, data, and workflows in real-world environments
These are the best entry points to understand how agentic systems operate in real-world environments.
-
BrandSense (Multi-Agent Brand Intelligence)
→ Brand analysis pipeline combining retrieval, scoring, and validation
→ Applies agent-based workflows to business evaluation scenarios -
Contract Risk Analysis (MCP + Foundry)
→ Contract analysis using MCP for tool-based evaluation and data access
→ Enables structured, repeatable, and auditable analysis workflows -
ITSM Multi-Agent System (Microsoft Foundry)
→ IT service management implemented with agents and structured orchestration
→ Covers ticket classification, routing, and lifecycle handling -
Agents Audit System
→ Observability and evaluation framework for agent/tool interactions
→ Tracks execution flow, decisions, and tool usage across workflows
These systems move beyond prompt-response interactions into structured, tool-driven execution.
Core components and architectural patterns used across agentic systems.
-
Semantic Kernel RAG System
→ Early RAG + orchestration pattern using Semantic Kernel and Azure services
→ Foundation for later agent-based system design -
Full-Stack RAG System
→ Retrieval + generation with React, FastAPI, and Cosmos DB -
Large Document Summarization (Durable Functions)
→ Distributed fan-out / fan-in processing for large-scale workflows -
Azure AI Foundry Deployment
→ Enterprise-ready infrastructure (networking, security, IaC) -
MCP YARP Gateway
→ Gateway layer for routing and controlling tool access in agentic systems
→ Enables structured interaction between agents and external tools
- Architecture + deep dives: https://stochasticcoder.com
- Connect with me: https://linkedin.com/in/jonathanscholtes/
All projects focus on:
- Real-world applicability
- System design over isolated prompts
- Production-oriented architectures