Autonomous systems / Knowledge book
Agentic AI
The agent isn’t one model anymore.
Connect decisions, tools, state, and human oversight.
The model proposes. The system decides what happens next.
Published in this topic
01Why an AI Agent Creates the Same Ticket Twice
Follow a lost tool response through an agent, policy gate, action ledger, and idempotent service. Learn why a timeout cannot prove an action failed.
The agent isn’t one model anymore.
Models propose. Tools execute. State remembers. The surrounding system makes the loop useful and accountable.
This collection grows section by section. Concepts describe the intended coverage; related reading may introduce only part of a section.
Start Here
Follow the numbered sections in order, or choose the subject you need. Published articles are linked where they support the section.
Concepts in this section
Follow the related reading where available. Dedicated guides for the remaining concepts are planned.
- What is an AI agent?The agent isn’t one model anymore.
- Agent vs chatbotThe agent isn’t one model anymore.
- Agent loopThe agent isn’t one model anymore.
Reasoning & Decisions
Concepts in this section
Follow the related reading where available. Dedicated guides for the remaining concepts are planned.
- ReasoningGuide planned
- PlanningGuide planned
- DecisionsGuide planned
- RoutingGuide planned
Related reading
No published reading is assigned to this section yet. Its place in the learning sequence is reserved as the collection grows.
Tool Calling
Concepts in this section
Follow the related reading where available. Dedicated guides for the remaining concepts are planned.
- FunctionsGuide planned
- APIsGuide planned
- ToolsThe agent isn’t one model anymore.
- Structured outputsGuide planned
State
Concepts in this section
Follow the related reading where available. Dedicated guides for the remaining concepts are planned.
- Session stateGuide planned
- Workflow stateGuide planned
- Persistent stateGuide planned
Memory
Concepts in this section
Follow the related reading where available. Dedicated guides for the remaining concepts are planned.
- Working memoryGuide planned
- Short-term memoryGuide planned
- Long-term memoryGuide planned
- Episodic memoryGuide planned
- Semantic memoryGuide planned
Related reading
No published reading is assigned to this section yet. Its place in the learning sequence is reserved as the collection grows.
Planning
Concepts in this section
Follow the related reading where available. Dedicated guides for the remaining concepts are planned.
- Task decompositionGuide planned
- PlanningGuide planned
- ReflectionGuide planned
- CritiqueGuide planned
Related reading
No published reading is assigned to this section yet. Its place in the learning sequence is reserved as the collection grows.
Agent Orchestration
Concepts in this section
Follow the related reading where available. Dedicated guides for the remaining concepts are planned.
- Sequential workflowsGuide planned
- Parallel workflowsGuide planned
- RoutersGuide planned
- SupervisorsGuide planned
Related reading
No published reading is assigned to this section yet. Its place in the learning sequence is reserved as the collection grows.
Multi-Agent Systems
Concepts in this section
Follow the related reading where available. Dedicated guides for the remaining concepts are planned.
- Specialized agentsGuide planned
- DelegationGuide planned
- CoordinationGuide planned
- CommunicationGuide planned
Related reading
No published reading is assigned to this section yet. Its place in the learning sequence is reserved as the collection grows.
Protocols
Concepts in this section
Follow the related reading where available. Dedicated guides for the remaining concepts are planned.
- MCPGuide planned
- A2AGuide planned
- Tool protocolsGuide planned
- Agent communicationGuide planned
Related reading
No published reading is assigned to this section yet. Its place in the learning sequence is reserved as the collection grows.
Human-in-the-Loop
Concepts in this section
Follow the related reading where available. Dedicated guides for the remaining concepts are planned.
- ApprovalsThe agent isn’t one model anymore.
- EscalationsGuide planned
- ReviewGuide planned
- FeedbackGuide planned
Agent Evaluation
Concepts in this section
Follow the related reading where available. Dedicated guides for the remaining concepts are planned.
- Task completionGuide planned
- Tool correctnessGuide planned
- Trajectory evaluationGuide planned
- Safety evaluationGuide planned
Related reading
No published reading is assigned to this section yet. Its place in the learning sequence is reserved as the collection grows.
Observability
Concepts in this section
Follow the related reading where available. Dedicated guides for the remaining concepts are planned.
- TracesGuide planned
- State inspectionGuide planned
- Tool logsGuide planned
- CostGuide planned
- LatencyGuide planned
Security
Concepts in this section
Follow the related reading where available. Dedicated guides for the remaining concepts are planned.
- PermissionsThe agent isn’t one model anymore.
- Tool accessGuide planned
- Prompt injectionGuide planned
- IsolationGuide planned
Production Agents
Concepts in this section
Follow the related reading where available. Dedicated guides for the remaining concepts are planned.
- ReliabilityGuide planned
- RetriesGuide planned
- IdempotencyGuide planned
- Long-running workflowsGuide planned
- Failure recoveryGuide planned