Insights on practical enterprise AI
Reference concepts and engineering questions for leaders and teams planning AI adoption, secure integration and responsible agentic workflows.
Reference
Key terms in agentic AI
- Agentic AI
-
Agentic AI describes systems in which AI agents pursue a goal by planning steps, using tools and acting on the results, within limits set by people.
- AI agent
-
An AI agent is software that uses a language model to decide what to do next and calls tools or business systems to carry it out.
- Multi-agent system
-
A multi-agent system divides a process among several specialized agents that coordinate with one another, each with its own tools and permissions.
- Retrieval-augmented generation (RAG)
-
RAG is a technique in which relevant passages are retrieved from an organization’s own content and supplied to a language model, so answers reflect that content and can cite it.
- Document intelligence
-
Document intelligence is the automated extraction, classification, comparison and summarization of information held in documents.
- Forward-deployed engineer
-
A forward-deployed engineer is embedded with a customer’s teams to build and deploy a solution inside the customer’s real operating environment.
- AI engineering enablement
-
Practical AI enablement can combine role-based workshops, architecture guidance and embedded engineering coaching.
- Cloud-native
-
Cloud-native software is designed for cloud platforms: packaged in containers, deployed by automated pipelines and able to scale with demand.
- Knowledge graph
-
A knowledge graph stores business entities together with the relationships between them, so questions about how things connect can be answered directly.
- Human-in-the-loop
-
Human-in-the-loop means a person reviews or approves an AI system’s proposed action before it takes effect, usually where the action has financial, regulatory or irreversible consequences.
- Agent orchestration
-
Agent orchestration is the coordination layer of a multi-agent system: it interprets a request, routes work to the right specialist agent, tracks workflow state and combines the results.
- Reference architecture
-
A reference architecture is a documented design that shows how a class of problem can be solved. It guides planning and sizing; it is not evidence of a deployed system.
Discuss your AI initiative
Tell us about your business workflow, users and existing systems. We will discuss the architecture, engineering support and next steps that fit your needs.