Document Processing · Enterprise Knowledge Search
Worley needed to unlock more than 50 years of engineering knowledge across over 100,000 documents, but fragmented tools, isolated repositories, and slow document retrieval caused engineers to waste time and made it hard to get accurate answers quickly.
Reported impact
Worley cut document-delivery and reading tasks from more than 15 minutes to under 5 minutes, is scaling toward 50,000 employees and 100,000-plus documents with 350 concurrent users, and targets 99.99 percent uptime for a secure knowledge platform that improves engineering productivity.
Worley built Worley AI.Assist as a unified engineering knowledge assistant on NVIDIA's enterprise AI stack. NVIDIA NIM microservices provide the layer for deploying and managing multiple LLMs and embedding models across Worley's global infrastructure, while NVIDIA Triton Inference Server lets the assistant serve several model types simultaneously with lower latency. For retrieval, NVIDIA NeMo Retriever converts both queries and source documents into vector embeddings so the system can find the most relevant engineering information, and a GPU-accelerated Milvus vector database performs similarity search across 100,000-plus document embeddings. Above that stack, Worley supports conversational search, uploaded-document question answering, temporary document snapshots, and agentic routing to specialized analysis tools, with results for some workflows stored in Microsoft Azure. Together, these layers turn a fragmented engineering corpus into a governed, scalable virtual assistant.
User questions or uploaded engineering files enter Worley AI.Assist, NeMo Retriever embeds both queries and documents, Milvus returns relevant knowledge-base matches, Triton and NIM serve the language and embedding models, and specialized document-analysis workflows can route results into Azure while the assistant returns cited answers to engineers.
At the heart of the platform are NVIDIA NIM™ microservices, which provide secure, production-ready environments for deploying generative AI models at scale.
For high-precision search and contextual understanding, Worley relies on NVIDIA NeMo™ Retriever.
Supporting this intelligence is a GPU-accelerated vector database, Milvus, which enables lightning-fast similarity search across more than 100,000 document embeddings.
This new solution, Worley AI.Assist, brings together all of Worley's engineering knowledge and best practices into a single, scalable virtual assistant.
What once took users more than 15 minutes ... can now take less than 5 minutes with Worley AI.Assist.
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
NVIDIA Sample Jul 7, 2026 Public Ministry of Rio Grande do Sul (MPRS) MPRS needed to manage rising legal caseloads, archived police inquiries, and public-service requests more efficiently, but the volume of legal documents and the need for compliant, locally controlled analysis made… 497 municipalities Archiving Intelligence Assistant (AIA) · Citizen Access Agent · Justice Intelligence platform NVIDIA Sample Feb 2, 2026 RBC Capital Markets RBC Capital Markets needed to speed up document-heavy research and banking workflows so analysts and bankers could deliver faster insights, prepare client materials more efficiently, and answer complex questions without… 10x increase in document processing capacity NVIDIA AI Enterprise · NVIDIA NIM · NVIDIA NeMo NVIDIA Sample Jan 17, 2025 National Informatics Centre NIC needed to reduce judicial backlog and improve accessibility in a court system with massive case volumes and many official languages, while giving judges faster ways to search and summarize legal material. 18,000 daily district court sessions NVIDIA AI Enterprise · NVIDIA DGX A100 Tensor Core GPUs · NVIDIA NIM 