Skip to content
NTPU AI Learn

AI 103 Apply

Design the memory, knowledge and skills of a real agent with a Canvas

What the Canvas is for

The Agent Memory, Knowledge & Skills Canvas is the capstone of AI 103. For one real use case you decide what the agent must know now, what it should remember, what it should retrieve, how knowledge is chunked and described, which reusable skills it needs, and where the boundaries are.

What you fill in

  • Agent Goal and Required Context
  • Memory: what persists across tasks, for how long, and who can access it
  • Knowledge Sources, and what to retrieve (RAG) rather than remember
  • Chunking Strategy and Metadata, including which metadata you will use as filters
  • Agent Skills: purpose, input, instructions, knowledge, tools, output and verification
  • Context Optimization: always include, retrieve when needed, summarize, compress or remove
  • Risks and Boundaries: what should not be stored, what the agent should not see, what needs access control or human review

Warning

A high-impact result should never be accepted without review. State which outputs a person must confirm before anyone relies on them.

Examples

The canvas page shows three provider-neutral examples: a customer support agent, a course material preparation agent, and a clinic administration summarization agent working on simulated documents.

Before you submit

Warning

Do not enter secrets, credentials, personal data, client data or sensitive information you are not authorized to share. Use simulated or public examples only.

What comes next: AI 104

Once an agent has memory, knowledge and skills, how do those abilities combine into a workflow that can run from start to finish? That is the question of AI 104: Agentic AI Workflow & Execution Design. As always, a human stays responsible for reviewing the result.

Go to the canvas

What you need to complete

  • Complete all six AI 103 modules
  • Score at least 80% on the course post-test
  • To earn the Agentic Memory Capability Certificate, also submit an Agent Memory, Knowledge & Skills Canvas and have it reviewed and passed by a Credential Admin against the rubric

How you are assessed

A Credential Admin scores the Agent Memory, Knowledge & Skills Canvas against six weighted domains: Agent Goal & Required Context, Memory Design, Knowledge Sources & Retrieval (RAG), Chunking & Metadata, Agent Skill Design, and Context Optimization, Risk & Human Review. A score of at least 15 out of 18, with no critical-fail rule triggered, is a PASS; 12 or below, or any triggered critical-fail rule, is a FAIL.

Info

This scoring logic is published so you know what is being assessed. It is not the same as revealing quiz answers.

Practical evidence

  • A recorded rubric result (PASS, REVISION_REQUIRED, or FAIL) with no critical-fail rule triggered
  • A PASS records capability evidence separately for each of the six fine-grained AI 103 competencies (Context vs. Memory Distinction, Memory Lifecycle Design, RAG & Retrieval Decisions, Chunking & Metadata Design, Knowledge Architecture & Agent Skills, Context Optimization & Failure Diagnosis)

What you can earn

AI 103 Course Completion Certificate: earned once you complete all six modules and meet the post-test threshold, with no Canvas submission required. Agentic Memory Capability Certificate: requires a separately submitted and reviewed Agent Memory, Knowledge & Skills Canvas. The two are independent and do not affect each other.

What this credential does not mean

This credential confirms the holder completed AI 103's required course and post-test, or (for the certificate) demonstrated memory design, knowledge retrieval, chunking, and skill design on the Canvas.

  • It does not mean the holder holds a formal knowledge-management or information-security professional license or is qualified for a specific role
  • It does not mean any specific agent memory system the holder designs later is automatically correct or safe
  • It does not mean human review can be skipped before a high-impact decision (Human, AI, Human)

What next?