This course teaches students to design, build, and deploy autonomous and semi-autonomous AI agents using modern, code-first frameworks, custom API-backed command-line tools, and the Model Context Protocol (MCP). Students will move from first-principles agent patterns (tool calling, ReAct, planning, memory) to production-grade multi-agent systems that integrate multiple LLM providers, package real APIs as agent-native CLIs, expose selected capabilities through MCP tools securely, and are instrumented for cost, latency, and observability. The course culminates in a capstone project that orchestrates a multi-agent system with CLI and MCP tool surfaces, human-in-the-loop checkpoints, and deployment guardrails.
54 contact hours (3-unit equivalent) over a 16-week term, approximately 3.4 hours per week. This is a non-credit course offered on the CDCP Certificate of Completion pathway.
AIINFRA 201 — students should be comfortable with Python, working with APIs, and basic cloud and deployment concepts before enrolling.
By the end of this course, students will be able to:
| Week | Topic |
|---|---|
| 01 | Agentic AI Foundations and the Provider Landscape |
| 02 | Tool Calling, ReAct, and Agent Planning Patterns |
| 03 | Memory and State Management for Agents |
| 04 | OpenAI, Claude, and Multi-Provider API Integration |
| 05 | AWS Bedrock, Google Vertex AI, and OpenRouter Routing |
| 06 | Cost, Latency, and Observability Instrumentation |
| 07 | Code-First Agent Frameworks I: LangGraph and Pydantic AI |
| 08 | Code-First Agent Frameworks II: CrewAI and OpenAI Agents SDK (Midterm) |
| 09 | MCP Architecture Deep Dive |
| 10 | Building Agent-Native CLI Tools and MCP Servers |
| 11 | MCP Authentication and Tool/Agent Security |
| 12 | Multi-Agent Orchestration and Handoffs |
| 13 | Human-in-the-Loop Workflows and Emerging Standards (A2A, AGENTS.md) |
| 14 | Agent Evaluation and Observability |
| 15 | Agent Deployment and Guardrails Introduction |
| 16 | Capstone Project & Course Review (Capstone) |
| Component | Weight |
|---|---|
| Labs | 40% |
| Discussions | 10% |
| Weekly Quizzes | 15% |
| Midterm | 15% |
| Final Capstone | 20% |
This course is graded Credit/No-Credit. A minimum of 70% overall is required to earn Credit.
Academic Integrity: Students are expected to submit their own work. Collaboration on concepts is encouraged, but submitted labs, quizzes, and the capstone project must reflect the student's individual understanding and implementation. Plagiarism or unauthorized sharing of graded solutions may result in disciplinary action consistent with district policy. Late Work: Assignments are due as posted in Canvas. Late submissions are accepted up to 7 calendar days after the due date with a notification to the instructor; work submitted after this window may not be eligible for credit except in documented emergencies. Responsible Use of AI: This course actively teaches and uses generative AI and agentic tools as its subject matter. Students may use AI assistants to support their learning and coding work, but must be able to explain, defend, and troubleshoot any code or agent behavior they submit. Using AI to fully generate a submission without understanding it undermines the learning outcomes of this course and is discouraged. Accessibility: This course complies with Section 508 and ADA accessibility standards. Students needing accommodations should contact the campus Disabled Students Programs and Services (DSPS) office as early as possible in the term so that appropriate arrangements can be made. Course materials are provided in accessible formats upon request.
All tools used in this course are free or available via free tier.
Notion ID: 392c08fd-0278-814b-a434-e79acd0faca1 Notion URL: https://app.notion.com/p/392c08fd0278814ba434e79acd0faca1