Highlights
- The Core Brain: Semantic Kernel & Function Logic
- Kernel Foundations (multi-model Kernel, DI, AI service structure)
- Native C# Plugins for AI Tool Execution
- Stepwise Planning with FunctionCallingStepwisePlanner
- Debugging Agent Failures and Reasoning Drift
- System Admin Agent Capstone Project
- The Digital Workforce: Multi-Agent Systems
- Agent Personas and System Instructions
- Orchestration Patterns and Agent Handoffs
- Memory and State Management with ChatHistory and Vector Stores
- Retrieval-Augmented Planning (RAP)
- Human-in-the-Loop Decision and Approval Workflows
- Content Pipeline Multi-Agent Project
- Production, Telemetry & .NET Aspire
- Aspire AppHost and Distributed Agent Architecture
Course Details
The Core Brain: Semantic Kernel & Function Logic
Focus: Building intelligent agents that use tools, plan steps, and reason iteratively.
1.1 Kernel Foundations
- Building a multi‑model Kernel using KernelBuilder
- Registering the Kernel via .NET Dependency Injection
- Structuring AI services for reasoning, summarization, and domain‑specific tasks
1.2 Native C# Plugins
- Writing C# classes as callable AI tools
- Using [KernelFunction] and [Description] to expose capabilities
- Passing strongly typed objects and records into LLM‑invoked functions
1.3 Stepwise Planning
- Implementing the FunctionCallingStepwisePlanner
- Understanding the reasoning loop: Goal → Plan → Tool Execution → Observation → Next Step
- Handling planner dead‑ends and misfires
1.4 Debugging Agent Failures
A new module focused on real‑world failure modes:
- Infinite loops
- Incorrect tool selection
- Planner hallucinations
- Bad arguments passed to functions
- Observing and correcting reasoning drift
Project — System Admin Agent
Build an agent that:
- Reads local logs
- Checks CPU usage via C# methods
- Diagnoses issues and suggests fixes
- Recovers gracefully from planner mistakes
The Digital Workforce: Multi‑Agent Systems
Focus: Designing specialized agents that collaborate, hand off tasks, and maintain memory.
2.1 Agent Personas & System Instructions
- Creating ChatCompletionAgent instances
- Defining backstories, constraints, and role boundaries
- Building personas such as:
- Researcher
- Writer
- Critic
2.2 Orchestration Patterns
- Sequential “waterfall” flows (Researcher → Writer → Critic)
- Group chat dynamics
- Agent handoffs when a task exceeds one agent’s expertise
- Designing workflows that mimic real digital teams
2.3 Memory & State
- Using ChatHistory for short‑term memory
- Implementing long‑term memory with vector stores (Azure AI Search, Qdrant)
- Retrieval‑Augmented Planning (new module):
- Agents retrieve relevant knowledge
- Use it to shape their plan
- Reduce hallucination and improve accuracy
2.4 Human‑in‑the‑Loop Patterns
- When agents should ask for clarification
- Approval workflows
- Escalation patterns for ambiguous or high‑risk tasks
Project — Content Pipeline
A multi‑agent workflow where:
- The Researcher gathers information
- The Writer produces a draft
- The Critic enforces brand voice and correctness
- The system can pause to ask the human for approval
Production, Telemetry & .NET Aspire
Focus: Making agents observable, reliable, and ready for real workloads.
3.1 Aspire AppHost
- Bootstrapping an Aspire solution
- Running agents, vector stores, and worker services under one orchestrated environment
- Service discovery without hardcoded URLs
3.2 Observability with OpenTelemetry
- Distributed tracing across multi‑agent workflows
- Visualizing each step of an agent’s reasoning loop
- Logging prompts and tool calls for debugging
- Identifying bottlenecks and failure points
3.3 Guardrails & Cost Control
- Token budgeting and loop‑kill conditions
- Output validation using FluentValidation
- Ensuring agents produce structured, schema‑compliant responses
Day 3 Project — Production Deployment
Deploy the Day 2 Content Pipeline into Aspire and observe:
- Agent reasoning steps
- Memory retrieval
- Token usage
- Planner decisions
- Human‑in‑the‑loop interactions
Capstone Project — The Knowledge Worker Agent (New)
A unified project spanning all three days.
Build an AI employee that can:
- Use C# plugins to perform real tasks
- Retrieve knowledge from vector memory
- Collaborate with other agents
- Ask the human for clarification or approval
- Run inside Aspire with full observability
- Stay within token and cost limits
Who should attend
- Software developers building AI-powered applications and agent systems
- .NET engineers looking to use Semantic Kernel and Azure AI in production
- Solution architects designing scalable, multi-agent AI architectures
- AI engineers and ML practitioners working on LLM-based systems
- Technical leads responsible for AI system design, deployment, and reliability
- Experienced developers interested in moving from LLM prototypes to production-grade AI solutions
Feedback
4.8 out of 5 average
"Our tailored course provided a well rounded introduction and also covered some intermediate level topics that we needed to know. " Brian Leek, Data Analyst, May 2022
“JBI did a great job of customizing their syllabus to suit our business. Our teams varied widely in terms of experience and the Instructor handled this particularly well - very impressive” Brian F, Team Lead, RBS, Data Analysis Course, 20 April 2024