Planning cockpit · becomes the syllabus

DRLXXX Agentic AI — Fundamentals and Applications

5 ECTS PhD seminar · Molde, Norway · Fall 2026 (tentative Week 35) · Swati Aggarwal & Bjørn Jæger · guest instruction by Jules White

DRLXXX Agentic AI — Fundamentals and ApplicationsFoundations of Applied AI for High School

The modules

BlockModule 1 · Module 1 · Prompts as the Foundation of Computation + Module 2 · Building AgentsModule 2 · Module 3 · Agent Tools, Knowledge, Memory & Research DesignModule 3 · Module 4 · Planning, Context & Multi-agent SystemsModule 4 · Module 5 · Supervised Research & Project DevelopmentModule 5 · Module 6 · Research Integration & Presentation
8:30–11:30
Module 1 · Prompts as the Foundation of ComputationAugmented Intelligence · 45mHow This Course Was Built · 15m✎Build a Prompt Pattern Library · 40mA Conversation Is One Big Prompt · 15mA Conversation Can Be Written Many Ways · 20mAn LLM Has No Memory · 15mConversations Are Plays: User, Assistant, System · 15m✎Exercise: You Are the Playwright · 25mNext-Word Prediction: Learning from Patterns · 15mTeaching by Example · 50m✎Exercise: Train by Example · 30mEvery Response Is Part of a Future Prompt · 20mAlignment: Training the Trajectory · 20m
Module 2 · Building AgentsWhat Is an AI Agent? · 25m✎Exercise: Your First Agents · 30mMultimodal Prompts: Images as Context · 20m✎Exercise: The Agent's Eyes · 25mTools: Constraining the Actions · 20mTool Parameters and Descriptions · 20m✎Exercise: Give the Agent Tools · 35mThe Conversation with the Computer · 20mProtocols and Function Calling · 25mShadows on the Wall · 20mThe Context Window: A Limit on the Conversation · 20mThe Economics of the Conversation · 20m✎Exercise: The Window Lab · 35mFraming: The Description Chooses the Solution · 25mThe Agent Loop · 30m✎Exercise: The Agent Forge · 45mWhere Does an Agent Get the Information It Needs? · 25m
765 / 180 min planned — over
Tools Are Words · 25mThe Language of Action · 30mTool Design Case Studies · 35m✎Exercise: The Design Studio · 40mSelf-Dialogue: Tools Made of Prompts · 25mWhat Comes Back: Designing Tool Results · 30mJust-in-Time Training: Results That Instruct · 30mErrors That Teach · 30mRead the Manual: Self-Training on Demand · 25mThe Table of Contents · 30m✎Exercise: The Archive · 30mSkills: The Brilliant Employee's First Day · 40m✎Exercise: Build a Skill · 40mThe Agent Teaches Its Future Self · 30mJumping-Off Points: Naming for the Searching Agent · 30m✎Exercise: The Relay · 45mThe One Percent: Errors That Detect · 30mCompaction: The Handoff to Yourself · 30mAbstraction: Briefing in Apples · 25m✎Hands-on Tool Design & Experimentation Searching by Meaning · 40mRetrieval Augmented Generation · 30mCorrect Information About the Wrong Thing · 25mGrounding: Asking the World If You're Right · 25mThe Questions a Tool Can Ask · 25mMemory: The Journal Between Conversations · 30mDiscovering Tools: The Catalog, Not the Toolbox · 25m✎Exercise: The Memory Lab · 45mAgentic Research Design Considerations
845 / 180 min planned — over
Module 3 Overview Agent Planning: The Language of Planning · 25mThinking in the Loop: Channels for Thought · 25mExecutable Plans: When the Plan Leaves the Conversation · 30mCognitive Load: The Two Loads · 30mLifetimes: What the Conversation Should Carry · 35mDo We Need an Agent at All? · 30mCode at the Forks: Workflows with Minds Inside · 30mThe Explorer · 30mThe Scientist and the Designer · 35mThe Planner–Executor · 30mThe Actor–Critic · 30mDesigning for Safety · 30mMulti-agent Systems: The Coordination of Conversations · 30mThe Architecture of Coordination · 25mThe Team: Role-Based Coordination · 30m✎Hands-on Multi-agent Systems Experimentation Task Decomposition: Trees with Evidence · 35mThe Task Is a Directory · 30m
510 / 180 min planned — over
Module 4 Overview Independent and Collaborative Research Project Development Applied Agentic AI Experimentation and Prototyping Faculty Supervision and Research Consultation Iterative Design and Methodological Refinement
0 / 180 min plannedIndependent & collaborative project development, faculty supervision
Module 5 Overview Final Research Project Development Session Guided Mentorship and Individual Feedback Collaborative Problem-Solving and Research Exchange Preparation of Research Demonstrations and Presentations
0 / 180 min plannedFinal project development, guided mentorship, demo preparation
12:30–13:45Agent design, prototyping & informal benchmarking (exercise: pattern-language)Agentic research project design session (exercise: human-as-runtime)Research project refinement & prototypingPeer feedback & collaborative project workshopsResearch project presentations
14:00–15:30Presentation of agent designs & scholarly discussionCollaborative discussion & presentation of research designsCollaborative discussion & presentation of research designsResearch progress presentations & scholarly discussionReflective discussion, course synthesis & research planning
rehearsed scripted outlined idea

Unscheduled modules

Backup9 lessons

Backlog

Every lesson, least finished first — this is the authoring worklist.

Augmented IntelligenceoutlinedModule 1 · Prompts as the Foundation of Computation · 45m
Exercise: The ArchiveoutlinedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 30m
Hands-on Tool Design & ExperimentationoutlinedModule 3 · Agent Tools, Knowledge, Memory & Research Design
Agentic Research Design ConsiderationsoutlinedModule 3 · Agent Tools, Knowledge, Memory & Research Design
Agentic Research Project Design SessionoutlinedModule 3 · Agent Tools, Knowledge, Memory & Research Design
Collaborative Discussion and Presentation of Research Project DesignsoutlinedModule 3 · Agent Tools, Knowledge, Memory & Research Design
Module 3 OverviewoutlinedModule 4 · Planning, Context & Multi-agent Systems
Hands-on Multi-agent Systems ExperimentationoutlinedModule 4 · Planning, Context & Multi-agent Systems
Module 4 OverviewoutlinedModule 5 · Supervised Research & Project Development
Independent and Collaborative Research Project DevelopmentoutlinedModule 5 · Supervised Research & Project Development
Applied Agentic AI Experimentation and PrototypingoutlinedModule 5 · Supervised Research & Project Development
Faculty Supervision and Research ConsultationoutlinedModule 5 · Supervised Research & Project Development
Iterative Design and Methodological RefinementoutlinedModule 5 · Supervised Research & Project Development
Peer Feedback and Collaborative Project WorkshopsoutlinedModule 5 · Supervised Research & Project Development
Research Progress Presentations and Scholarly DiscussionoutlinedModule 5 · Supervised Research & Project Development
Module 5 OverviewoutlinedModule 6 · Research Integration & Presentation
Final Research Project Development SessionoutlinedModule 6 · Research Integration & Presentation
Guided Mentorship and Individual FeedbackoutlinedModule 6 · Research Integration & Presentation
Collaborative Problem-Solving and Research ExchangeoutlinedModule 6 · Research Integration & Presentation
Preparation of Research Demonstrations and PresentationsoutlinedModule 6 · Research Integration & Presentation
Research Project PresentationsoutlinedModule 6 · Research Integration & Presentation
Reflective Discussion on Agentic AI Research Directions and Future ApplicationsoutlinedModule 6 · Research Integration & Presentation
Course Synthesis and Research PlanningoutlinedModule 6 · Research Integration & Presentation
What is Agentic AI?outlinedBackup · 25m
Hands-on Agent Planning & PrototypingoutlinedBackup
Prompt Engineering for AgentsoutlinedBackup · 30m
Hands-on Agent Cognitive Load Management ExperimentationoutlinedBackup
Hands-on Human-in-the-Loop Agent PrototypingoutlinedBackup
What Is Agentic AI? + Prompt Engineering Fundamentals for AgentsoutlinedBackup
Hands-on Agent Skill DesignoutlinedBackup
Hands-on Agent Rapid Prototyping & ExperimentationoutlinedBackup
Hands-on Multi-agent Task Decomposition ExperimentationoutlinedBackup
How This Course Was BuiltscriptedModule 1 · Prompts as the Foundation of Computation · 15m
Build a Prompt Pattern LibraryscriptedModule 1 · Prompts as the Foundation of Computation · 40m
The Research Arc of the Week: From Hands-On Agents to a PaperscriptedModule 1 · Prompts as the Foundation of Computation · 15m
A Conversation Is One Big PromptscriptedModule 1 · Prompts as the Foundation of Computation · 15m
A Conversation Can Be Written Many WaysscriptedModule 1 · Prompts as the Foundation of Computation · 20m
An LLM Has No MemoryscriptedModule 1 · Prompts as the Foundation of Computation · 15m
Conversations Are Plays: User, Assistant, SystemscriptedModule 1 · Prompts as the Foundation of Computation · 15m
Exercise: You Are the PlaywrightscriptedModule 1 · Prompts as the Foundation of Computation · 25m
Next-Word Prediction: Learning from PatternsscriptedModule 1 · Prompts as the Foundation of Computation · 15m
Teaching by ExamplescriptedModule 1 · Prompts as the Foundation of Computation · 50m
Exercise: Train by ExamplescriptedModule 1 · Prompts as the Foundation of Computation · 30m
Every Response Is Part of a Future PromptscriptedModule 1 · Prompts as the Foundation of Computation · 20m
Alignment: Training the TrajectoryscriptedModule 1 · Prompts as the Foundation of Computation · 20m
What Is an AI Agent?scriptedModule 2 · Building Agents · 25m
Exercise: Your First AgentsscriptedModule 2 · Building Agents · 30m
Multimodal Prompts: Images as ContextscriptedModule 2 · Building Agents · 20m
Exercise: The Agent's EyesscriptedModule 2 · Building Agents · 25m
Tools: Constraining the ActionsscriptedModule 2 · Building Agents · 20m
Tool Parameters and DescriptionsscriptedModule 2 · Building Agents · 20m
Exercise: Give the Agent ToolsscriptedModule 2 · Building Agents · 35m
The Conversation with the ComputerscriptedModule 2 · Building Agents · 20m
Protocols and Function CallingscriptedModule 2 · Building Agents · 25m
Shadows on the WallscriptedModule 2 · Building Agents · 20m
The Context Window: A Limit on the ConversationscriptedModule 2 · Building Agents · 20m
The Economics of the ConversationscriptedModule 2 · Building Agents · 20m
Exercise: The Window LabscriptedModule 2 · Building Agents · 35m
Framing: The Description Chooses the SolutionscriptedModule 2 · Building Agents · 25m
The Agent LoopscriptedModule 2 · Building Agents · 30m
Exercise: The Agent ForgescriptedModule 2 · Building Agents · 45m
Where Does an Agent Get the Information It Needs?scriptedModule 2 · Building Agents · 25m
Agent Design, Prototyping, and Informal Benchmarking SessionscriptedModule 2 · Building Agents · 90m
The Day After: Predicting a Research OpportunityscriptedModule 2 · Building Agents · 90m
Positioning: How It's Done Today — and the Research Opportunity for TomorrowscriptedModule 2 · Building Agents · 60m
Tools Are WordsscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 25m
The Language of ActionscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 30m
Tool Design Case StudiesscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 35m
Exercise: The Design StudioscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 40m
Self-Dialogue: Tools Made of PromptsscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 25m
What Comes Back: Designing Tool ResultsscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 30m
Just-in-Time Training: Results That InstructscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 30m
Errors That TeachscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 30m
Read the Manual: Self-Training on DemandscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 25m
The Table of ContentsscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 30m
Skills: The Brilliant Employee's First DayscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 40m
Exercise: Build a SkillscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 40m
The Agent Teaches Its Future SelfscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 30m
Jumping-Off Points: Naming for the Searching AgentscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 30m
Exercise: The RelayscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 45m
The One Percent: Errors That DetectscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 30m
Compaction: The Handoff to YourselfscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 30m
Abstraction: Briefing in ApplesscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 25m
Build It and Try It: Exploring Agent Designs for Your DomainscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 75m
Searching by MeaningscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 40m
Retrieval Augmented GenerationscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 30m
Correct Information About the Wrong ThingscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 25m
Grounding: Asking the World If You're RightscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 25m
The Questions a Tool Can AskscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 25m
Memory: The Journal Between ConversationsscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 30m
Discovering Tools: The Catalog, Not the ToolboxscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 25m
Exercise: The Memory LabscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 45m
Rough Proof of Concept: Trying Your Agent for RealscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 75m
How Will You Know? — Designing the Study You'll RunscriptedModule 3 · Agent Tools, Knowledge, Memory & Research Design · 90m
Agent Planning: The Language of PlanningscriptedModule 4 · Planning, Context & Multi-agent Systems · 25m
Thinking in the Loop: Channels for ThoughtscriptedModule 4 · Planning, Context & Multi-agent Systems · 25m
Executable Plans: When the Plan Leaves the ConversationscriptedModule 4 · Planning, Context & Multi-agent Systems · 30m
Cognitive Load: The Two LoadsscriptedModule 4 · Planning, Context & Multi-agent Systems · 30m
Lifetimes: What the Conversation Should CarryscriptedModule 4 · Planning, Context & Multi-agent Systems · 35m
Do We Need an Agent at All?scriptedModule 4 · Planning, Context & Multi-agent Systems · 30m
Code at the Forks: Workflows with Minds InsidescriptedModule 4 · Planning, Context & Multi-agent Systems · 30m
The ExplorerscriptedModule 4 · Planning, Context & Multi-agent Systems · 30m
The Scientist and the DesignerscriptedModule 4 · Planning, Context & Multi-agent Systems · 35m
The Planner–ExecutorscriptedModule 4 · Planning, Context & Multi-agent Systems · 30m
The Actor–CriticscriptedModule 4 · Planning, Context & Multi-agent Systems · 30m
Designing for SafetyscriptedModule 4 · Planning, Context & Multi-agent Systems · 30m
Multi-agent Systems: The Coordination of ConversationsscriptedModule 4 · Planning, Context & Multi-agent Systems · 30m
The Architecture of CoordinationscriptedModule 4 · Planning, Context & Multi-agent Systems · 25m
The Team: Role-Based CoordinationscriptedModule 4 · Planning, Context & Multi-agent Systems · 30m
Task Decomposition: Trees with EvidencescriptedModule 4 · Planning, Context & Multi-agent Systems · 35m
The Task Is a DirectoryscriptedModule 4 · Planning, Context & Multi-agent Systems · 30m
The Skeleton Assembly: Building Paper Skeleton v1scriptedModule 4 · Planning, Context & Multi-agent Systems · 90m
Agentic Research Project Refinement & Prototyping SessionscriptedModule 4 · Planning, Context & Multi-agent Systems · 75m
Collaborative Discussion and Presentation of Research Project DesignsscriptedModule 4 · Planning, Context & Multi-agent Systems · 90m