Skip to main content
Version: Next

Agent Architecture

1. What the architecture does

Prodia Agent architecture is responsible for turning a user’s industrial question into a controlled execution path and an actionable response.

2. Why one entry point does not mean one monolithic Agent

From the user’s perspective, Prodia can appear as one conversational Agent entry point. Internally, however, it is better understood as an orchestrated system that routes work across scenario logic, governed capabilities, context handling, memory, and knowledge.

That means the same conversational entry point may organize different kinds of capability according to the question, such as:

  • production and business analysis
  • quality and anomaly diagnosis
  • fault and equipment investigation
  • knowledge retrieval and experience enhancement
  • result organization and explanation

The value of this design is that the system does not ask one monolithic Agent to do everything equally well. It lets the right capability participate in the right scenario.

3. Logical layers

LayerResponsibilityUser value
Interaction layerreceive natural-language questions and support follow-up turnsusers can keep analyzing through conversation
Routing and orchestration layeridentify intent, scope, and execution pathcomplex questions become structured steps
Capability module layerapply domain methods and usage rules for different roles and scenariosresponses stay closer to operational reality
Scenario capability layerorganize output, quality, takt, fault, SPC, and other domain capabilityanalysis stays grounded in industrial language
Memory layerpreserve continuity, preference, and intermediate task statefewer repeated clarifications across turns
Tool execution layerinvoke governed business tools and structured capabilitiesanswers stay grounded in real execution
Knowledge enhancement layerbring in cases, rules, glossary, and organizational experienceimproves explanation and reuse
Governance layerkeep permissions, safety, and trust boundaries intactthe system stays reliable and controllable

4. Why the architecture is now better explained as three collaborative layers

If we look deeper into runtime behavior, the current Prodia architecture is no longer best explained as “prompt-first”. It is better understood through three layers:

LayerFocusWhat it does
Harness Engineeringruntime, tools, sessions, events, state, and governancedefines the execution world in which Agent operates
Context Engineeringhistory, memory, knowledge, time, tool boundary, and budgetdecides what this turn is allowed to see
Prompt Engineeringrole, rules, mode switch, skill visibility, and output contractdecides how the turn should behave and respond

In simple terms:

  • Harness builds the execution world
  • Context selects what the turn can see
  • Prompt shapes how the turn behaves and explains

For a deeper explanation, continue with Three-layer Architecture, Context Engineering, and Prompt Engineering.

5. Relationship overview

Seen through the three-layer lens:

  • Harness guarantees that sessions, tools, state, and execution boundaries really exist
  • Context decides how session context, memory, and knowledge enter the turn
  • Prompt turns those inputs into stable behavior and output constraints

6. Architectural principle

The Agent should not behave like a free-form generator.

Its job is to organize reliable industrial execution under context and boundary, then turn the result into something operations teams can use immediately.