MST 501Mastery · 90 days3 US credit hours·Mastery course · Upper-division / professional elective·v1 · updated 3 Oct 2026

AI Architecture Mastery

University / Board Instructor Student Outcomes & Skills

Course outcomes

CO1

Demonstrate Foundations & Agent Architecture competence: complete the Day 1–15 builds, explain the underlying theory, and defend the phase artefact under live questioning.

C2 Understand

Agent architecture: the four execution zones and ten layers Building an agent loop from primitives (no framework)Workflow vs. agent decision-making and control selection Structured outputs & schema validation Reliability plumbing: timeouts, retries, idempotency, circuit breakers Evaluation engineering with golden sets AI security & prompt-injection defence Cost and latency optimisation against a quality gate Failure taxonomy, improvement flywheel, labelling operations and incident response

CO2

Demonstrate Model Layer, Context & Tools competence: complete the Day 16–30 builds, explain the underlying theory, and defend the phase artefact under live questioning.

C3 Apply

Structured outputs & schema validation Context engineering Prompt engineering Model selection, routing, fallback and provider portability Tool contract design, descriptions and token-efficient returns Read/write separation, propose-commit and dry-run patterns Reliability plumbing: timeouts, retries, idempotency, circuit breakers Graph orchestration with typed state and conditional control flow

CO3

Demonstrate Knowledge, Memory & Retrieval competence: complete the Day 31–45 builds, explain the underlying theory, and defend the phase artefact under live questioning.

C4 Analyse

Context engineering RAG pipeline design Hybrid search, chunking & reranking Metadata design, tenant isolation and clearance-aware filtering Query understanding: rewriting, decomposition, adaptive retrieval Citations, groundedness and the ten-cause RAG diagnosis

CO4

Demonstrate MCP, Orchestration & Multi-Agent competence: complete the Day 46–60 builds, explain the underlying theory, and defend the phase artefact under live questioning.

C4 Analyse

Tool contract design, descriptions and token-efficient returns Reliability plumbing: timeouts, retries, idempotency, circuit breakers Tool calling & MCP integration MCP authorization, scoping and enterprise hardening A2A: agent cards, delegation and cross-framework interoperability Graph orchestration with typed state and conditional control flow Multi-agent orchestration

CO5

Demonstrate Evaluation, Observability & Security competence: complete the Day 61–75 builds, explain the underlying theory, and defend the phase artefact under live questioning.

C5 Evaluate

Read/write separation, propose-commit and dry-run patterns MCP authorization, scoping and enterprise hardening LLMOps observability & SLOs Evaluation engineering with golden sets CI regression gates for AI AI security & prompt-injection defence AI red-teaming

CO6

Demonstrate Production, Operations & Mastery competence: complete the Day 76–90 builds, explain the underlying theory, and defend the phase artefact under live questioning.

C6 Create

Agent architecture: the four execution zones and ten layers Workflow vs. agent decision-making and control selection Context engineering Graph orchestration with typed state and conditional control flow CI regression gates for AI AI red-teaming Containerisation, secrets management and CI/CD for agents Staged rollout, kill switches, scaling and backpressure Cost and latency optimisation against a quality gate Failure taxonomy, improvement flywheel, labelling operations and incident response Responsible AI & regulatory compliance Architecture review, framework selection and teaching the discipline

Phases & skills

Phase	Topic	Outcomes	Deliverable	Skills
1	Foundations & Agent Architecture Days 1–15 · C2	CO1	CAPSTONE (Phase 1) — the frameworkless structured analyst	Agent architecture: the four execution zones and ten layers Building an agent loop from primitives (no framework)Workflow vs. agent decision-making and control selection Structured outputs & schema validation Reliability plumbing: timeouts, retries, idempotency, circuit breakers Evaluation engineering with golden sets AI security & prompt-injection defence Cost and latency optimisation against a quality gate Failure taxonomy, improvement flywheel, labelling operations and incident response
2	Model Layer, Context & Tools Days 16–30 · C3	CO2	CAPSTONE (Phase 2) — the tool-using support agent with approvals	Structured outputs & schema validation Context engineering Prompt engineering Model selection, routing, fallback and provider portability Tool contract design, descriptions and token-efficient returns Read/write separation, propose-commit and dry-run patterns Reliability plumbing: timeouts, retries, idempotency, circuit breakers Graph orchestration with typed state and conditional control flow
3	Knowledge, Memory & Retrieval Days 31–45 · C4	CO3	CAPSTONE (Phase 3) — the tenant-aware enterprise policy analyst	Context engineering RAG pipeline design Hybrid search, chunking & reranking Metadata design, tenant isolation and clearance-aware filtering Query understanding: rewriting, decomposition, adaptive retrieval Citations, groundedness and the ten-cause RAG diagnosis
4	MCP, Orchestration & Multi-Agent Days 46–60 · C4	CO4	CAPSTONE (Phase 4) — the durable, human-approved transaction agent	Tool contract design, descriptions and token-efficient returns Reliability plumbing: timeouts, retries, idempotency, circuit breakers Tool calling & MCP integration MCP authorization, scoping and enterprise hardening A2A: agent cards, delegation and cross-framework interoperability Graph orchestration with typed state and conditional control flow Multi-agent orchestration
5	Evaluation, Observability & Security Days 61–75 · C5	CO5	CAPSTONE (Phase 5) — the evaluated, instrumented, hardened agent	Read/write separation, propose-commit and dry-run patterns MCP authorization, scoping and enterprise hardening LLMOps observability & SLOs Evaluation engineering with golden sets CI regression gates for AI AI security & prompt-injection defence AI red-teaming
6	Production, Operations & Mastery Days 76–90 · C6	CO6	MASTERY CAPSTONE II — measure it against the rubric	Agent architecture: the four execution zones and ten layers Workflow vs. agent decision-making and control selection Context engineering Graph orchestration with typed state and conditional control flow CI regression gates for AI AI red-teaming Containerisation, secrets management and CI/CD for agents Staged rollout, kill switches, scaling and backpressure Cost and latency optimisation against a quality gate Failure taxonomy, improvement flywheel, labelling operations and incident response Responsible AI & regulatory compliance Architecture review, framework selection and teaching the discipline