Agents and agentic workflows

Agentic systems designed, connected, kept running, measured and constrained: agent architectures, MCP (both angles), durable orchestration, evaluation, guardrails, LLMOps.

This module covers how agentic systems are designed, connected to other systems, kept running through failures, measured, and constrained. It moves from agent architectures, frameworks and SDK primitives through the Model Context Protocol (the concepts and the building of MCP servers) and durable orchestration to evaluation, guardrails and LLMOps, which includes evals, tracing and AI-assisted development with Claude Code and Cursor. These are the parts that decide whether an agentic deployment survives contact with production.

Lessons (7)

• Agent architectures, agent frameworks and the Model Context Protocol · Practice

  An agent is a model that decides its own next step in a loop of reasoning, tool calls and observations, as opposed to a workflow whose steps are fixed in code.

• LLMOps: evals, tracing, guardrails and AI-assisted development with Claude Code and Cursor · Practice

  LLMOps applies the release and operations discipline of MLOps to systems whose behavior is set by prompts, models, retrieval and tools together.

• Agent architectures and agentic workflows with agent SDKs

  An agentic system sits somewhere between a fixed workflow, where code decides the order of model calls, and an autonomous agent, where the model decides its next step in a loop with tools.

• Building Model Context Protocol (MCP) servers · Practice

  The Model Context Protocol is an open standard for connecting AI applications to external systems, so one integration works across many hosts instead of being rewritten per agent framework.

• Agent orchestration and durable execution · Practice

  Orchestration decides which agent or step runs next and what context it receives; durable execution makes sure a long run survives crashes, deploys and rate limits without losing or repeating work.

• Evaluating LLM and agent systems · Practice

  Evaluation turns a vague sense that the system seems fine into a number tied to defined success criteria, so prompts, models and architectures can be changed with evidence.

• Responsible AI controls and guardrails · Practice

  Responsible AI work combines a governance process that identifies and tracks risk with technical controls that limit what a model-driven system can see, say and do.

Changes

• Version 1, 2026-10-14: re-cut from the role pathway groups into a topic module (site 2026.10.2); no lesson changed