Agentic AI Reference Architecture

Knowledge & Data · Software component

Context Assembler

Software componentKnowledge & DataKnowledge & DataVariation point (abstract)arc:ContextAssembler

A software component that combines retrieved document content and graph relationship context into a grounded prompt context.

Responsibility. Builds grounded prompt context from documents and graph relations.

Also known as: Hybrid Context Builder, Context optimization

sends data to; is invoked byinvokes; receives data fromreceives data fromis orchestrated bysends data toreceives data frominvokesinvokesis configured byreceives data fromis specialized byreceives data fromAnswer Synthesizer: sends data to; is invoked byAnswer SynthesizerContext Compressor: invokes; receives data fromContext CompressorRetriever: receives data fromRetrieverRAG Query Orchestrator: is orchestrated byRAG Query OrchestratorPrompt Context Builder: sends data toPrompt Context BuilderReranker: receives data fromRerankerReasoning Consistency Checker: invokesReasoning Consistency Ch…Dense-Sparse Hybrid Retriever: invokesDense-Sparse Hybrid Retr…Knowledge Chunk Metadata Schema: is configured byKnowledge Chunk Metadata…Parallel Fusion Retriever: receives data fromParallel Fusion RetrieverMultimodal Context Assembler: is specialized byMultimodal Context Assem…Unified Embedding Retriever: receives data fromUnified Embedding Retrie…
Direct neighbourhood (hover for relationship types)

Variants

VariantWhen to choose
Multimodal Context Assembler—

Relationships

is configured by structural

invokes dependency

is invoked by dependency

receives data from dynamic

sends data to dynamic

is orchestrated by control

Design guidance

Quantitative guidance

As stated by the sources; verify before use.

Classification

Patterns
Attention-aware positioning: most important item first, second-most important last, tertiary items in the middle
Quality attributes
Functional suitability: correctness and validity (ISO/IEC 25010 | NIST AI RMF: valid)
Risks mitigated
Lost-in-the-middle position biasOversized context windows tripling token cost

Sources

  1. Ch1.7B: T. Nguyen, "Relational Reasoning with Knowledge Graphs - Hybrid RAG+KG Integration," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 1.7B. ISBN: 9798244538229.
  2. Ch2.7: T. Nguyen, "Multimodal RAG Approaches," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 2.7. ISBN: 9798244538229.
  3. Ch2.9: T. Nguyen, "Streaming and Real-Time Responses," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 2.9. ISBN: 9798244538229.
  4. Ch3.10: T. Nguyen, "Efficiency Metrics," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 3.10. ISBN: 9798244538229.
  5. Ch5.9: T. Nguyen, "Working Memory," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 5.9. ISBN: 9798244538229.
  6. Ch5.13: T. Nguyen, "Hybrid Decision Systems Fundamentals," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 5.13. ISBN: 9798244538229.
  7. Ch6.2B: T. Nguyen, "Production Vector Database Deployment," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 6.2B. ISBN: 9798244538229.
  8. Ch6.5: T. Nguyen, "Production RAG Systems," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 6.5. ISBN: 9798244538229.