About the Role

Purple Fabric is looking for a Knowledge Architect to own the architecture and evolution of its enterprise knowledge plane — the foundation that enables AI agents and products to understand, retrieve and use enterprise information reliably.

You will shape how information is ingested, processed, represented, secured, retrieved, evaluated and reused across Purple Fabric and its related platforms.

This is a hands-on architecture role combining enterprise knowledge systems, RAG, semantic and graph technologies, data governance and production engineering. You will work closely with product and engineering teams to turn complex knowledge requirements into scalable, reusable platform capabilities.

What You Will Own

1. Knowledge-Plane Architecture

- Define the architecture, boundaries and interfaces across ingestion, parsing, OCR, chunking, metadata, embeddings, vector search, semantic and graph structures, retrieval, reranking and citations.
- Guide the evolution of the knowledge plane across enterprise AI use cases.

2. Knowledge Representation

- Design flexible models for documents, entities, relationships, provenance and semantic context.
- Determine when vector, relational, graph or hybrid approaches are appropriate.

3. Retrieval Quality

- Define architecture for RAG, hybrid search, self-query, reranking and context assembly.
- Establish measurable retrieval and evaluation criteria to assess knowledge coverage, relevance, completeness and answer quality.

4. Knowledge Lifecycle and Governance

- Own versioning, refresh, deletion, lineage, access-control propagation and tenant isolation.




- Address data residency, retention and policy enforcement throughout the knowledge lifecycle.

5. Scale, Performance and Cost

- Define measurable targets for ingestion throughput, indexing and search latency, storage growth, retrieval quality and cost.
- Guide benchmarking, capacity planning and performance optimisation.

6. Reusable Platform Capabilities

- Translate recurring Purple Fabric, MSOCK and customer requirements into reusable knowledge capabilities and adapters.
- Support integrations with customer-managed and external knowledge stores where appropriate.

7. Proofs of Concept and Technical Execution

- Prototype high-risk knowledge architecture decisions.
- Review implementations, troubleshoot retrieval and data issues, and use evidence to guide what should be built, refined or retired.

What You Bring

- Production knowledge and RAG architecture: Experience architecting a live enterprise knowledge, search, RAG or semantic platform, with the ability to explain actual data volumes, query characteristics, quality metrics and design decisions.
- End-to-end ingestion and retrieval expertise: Strong understanding of document processing, metadata, chunking, embeddings, indexing, retrieval, reranking,



citations and real-world failure modes.
- Semantic and graph modelling: Ability to model entities, relationships, ontologies, semantic structures and provenance, and understand when graph-based approaches add value.
- Evaluation discipline: Experience defining coverage, retrieval relevance, grounding, completeness and answer-quality evaluations beyond subjective demonstrations.
- Enterprise governance: Understanding of access controls, tenant isolation, PII, lineage, retention, auditability and customer-hosted or data-residency requirements.
- Hands-on engineering: Ability to prototype using Python, Java or equivalent, review APIs and data models, troubleshoot pipelines and work directly with engineering teams on production issues.

Valuable to Have

- Experience with MongoDB Atlas Search, OpenSearch, Elasticsearch, pgvector, Azure AI Search, Neo4j or other vector, graph and search technologies.
- Exposure to code and software knowledge graphs, ontology-driven systems or engineering knowledge platforms.
- Experience with multi-cloud, customer-hosted and regulated-enterprise deployments.
- Familiarity with agentic AI, knowledge-tool integration, evaluation frameworks and AI-assisted engineering.

Interested?

If you have built and evolved enterprise knowledge platforms in production and want to shape how AI systems access and use enterprise knowledge, we'd like to hear from you.

Apply through this LinkedIn job listing or contact the recruiter listed in the accompanying recruitment post.

📌 Knowledge Architect (Chennai)
🏢 Intellect Design Arena
📍 Chennai