Reported impact
Business outcome
Cohesity reports better retrieval recall, fewer missed relevant documents, reduced model confusion, and an improved user experience for Gaia customers.
- Customer Experience
- Productivity Improvement
NVIDIASoftware / SaaSEnterprise Knowledge SearchAI and Machine LearningGenerative AI / LLMCustomer ExperienceProductivity ImprovementUndated NVIDIANV#76b900Software / SaaSUnited StatesCohesity needed to improve retrieval accuracy in its generative AI assistant so relevant enterprise documents were not missed before answer generation, especially in high-stakes use cases such as legal eDiscovery.Gaia’s errors were reduced by 13%.Est. · Dec 8, 20245Enterprise Knowledge SearchCUDA Python · NVIDIA A100 Tensor Core GPUs · NVIDIA NIM microservices0 2024-12-08 5
Cohesity
Enterprise Knowledge Search
Cohesity needed to improve retrieval accuracy in its generative AI assistant so relevant enterprise documents were not missed before answer generation, especially in high-stakes use cases such as legal eDiscovery.
Gaia’s errors were reduced by 13%.
Solution
Cohesity uses NVIDIA technologies inside Cohesity Gaia, its generative AI-powered conversational search assistant for enterprise data. Gaia is described as having two core stages—search/retrieval and generation—and Cohesity inserted the NVIDIA NeMo Retriever text reranking NIM microservice into the retrieval path to improve which documents are passed downstream. Queries first retrieve candidate documents from combined enterprise data sources; the reranking service then scores each document-query pair, reorders the results by relevance, and forwards only the strongest context to the generation step. Cohesity says this reduces irrelevant context and model confusion while improving recall. The deployment is supported by broader NVIDIA infrastructure that Cohesity already uses, including NVIDIA Triton Inference Server, NVIDIA A100 Tensor Core GPUs, and CUDA Python. NVIDIA NIM supplies the containerized microservice packaging and deployment model Cohesity uses to operationalize inference across environments.
Data flow
A user query enters Cohesity Gaia, which retrieves candidate documents from multiple enterprise data sources. The NVIDIA NeMo Retriever text reranking NIM microservice assigns relevance scores to each document-query pair and reorders the set so only the most relevant documents are passed to Gaia's generation stage for summarization or answer creation.
Solution architecture
- Compute
- NVIDIA NeMo Retriever text reranking NIM microservice Scores and reorders retrieved documents so Gaia sends the most relevant context to generation.
- NVIDIA A100 Tensor Core GPUs Provide accelerated compute for Gaia's AI workloads.
- CUDA Python Supports GPU-accelerated AI application development in Cohesity's implementation.
- Serving
- NVIDIA Triton Inference Server Supports AI model inference in Cohesity's NVIDIA-based stack.
- Uncategorized
- NVIDIA NIM microservices Provide the containerized microservices framework Cohesity uses to deploy optimized AI inference components across environments.
Architecture clues
- Cohesity Gaia has two important components: search/retrieval and generation.
- NIM microservices are containerized and designed for deployment across clouds, data centers, and workstations.
- The pipeline combines multiple sources of data.
- The reranking service is inserted between initial retrieval and the subsequent generation step.
Evidence from the source
"Cohesity Gaia has two important components: search/retrieval and generation. The NVIDIA NeMo Retriever text reranking NIM microservice increases the recall performance of the retrieval component."
"Earlier this year, we introduced the industry’s first generative AI-powered conversational search assistant, Cohesity Gaia. It brings the power of generative AI to enterprise data"
"It works by reordering retrieved documents based on their relevance to a user query, assigning a 'relevance score' to each document-query pair."
"Our most extensive NIM implementation in Cohesity Gaia is with the NeMo Retriever text reranking NIM microservice."
"Thanks to the use of text reranking NIM microservice, Gaia’s errors were reduced by 13%."
"the text reranking NIM microservice will then refine this list by scoring and reordering them to ensure the most relevant documents are only sent to the subsequent generation step"
Transparency Evidence & methodology
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
How scores are calculated
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Customer reference value85/100
- Problem clarity20/20Measures whether the customer's operational problem is specific enough to compare.
- Implementation depth20/25Rewards named components, data flow, integrations, and architecture detail.
- Outcome specificity20/20Measures whether the reported benefit is concrete rather than a generic claim.
- Quantified impact15/251 metric describe measurable impact.
- Deployment scale10/10Rewards evidence of a live or scaled deployment and meaningful integration scope.
Evidence confidence89%
- Source authority25/25The deployment is described by the adopting organization or a government source.
- Claim traceability25/2512 of 12 key fields link to a named source.
- Direct evidence20/206 direct source quotes retained.
- Date provenance0/10The publication date has no documented provenance.
- Source transparency9/102 source references document publisher, URL, authority, retrieval, and supported fields.
- Implementation specificity10/10Rewards source-backed components, data flow, quantified impact, and a specific outcome.
Traceability
Sources & provenance
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
- https://www.cohesity.com/blogs/how-cohesity-uses-nvidia-nemo-retriever-microservices-to-improve-rag-ai-retrieval-recall/CohesityFirst-party
Cohesity customer story
Supports- Customer
- Business problem
- Solution
- Products
- Solution components
- Data flow
- Integrations
- technical area
- use case category
- Business outcome
- outcome category
- Metrics
- Architecture
- evidence quotes
- https://builtin.com/company/cohesityBuilt InIndependent secondary
Cohesity Careers, Perks + Culture | Built In
The Organization JSON-LD lists streetAddress '300 Park Ave', addressLocality 'San Jose', addressCountry 'US', and numberOfEmployees value 1529.
Supports- Company size
- Region
Date evidence
- Unverified extracted date Dec 8, 2024 · Cohesity · Unknown Legacy extracted date without source-page confirmation.Open date source
Not separately sourced Industry. Treat these normalized attributes as lower-confidence until a dedicated source is added.
Related
Similar use cases
NVIDIA Sample Jul 7, 2026 Siam AI Corporation Co., Ltd. (Siam.AI Cloud) Thailand tourism information was fragmented across multiple websites, difficult to verify, often outdated, and hard for visitors to access across languages, creating poor experiences and inefficient support workflows. NVIDIA AI Enterprise · NVIDIA NIM · NVIDIA NeMo NVIDIA Sample Feb 2, 2026 NVIDIA NVIDIA needed to move employee-built chatbots and agents out of shadow-AI pilots and onto a unified, secure, high-performance platform that could support hundreds of governed AI workflows across engineering, software… from 15 hours to one hour NVIDIA AI Blueprint for Retrieval-Augmented Generation (RAG) · NVIDIA AI Enterprise · NVIDIA GPU Operator NVIDIA Sample Aug 13, 2025 Perplexity Perplexity needed to customize frontier models for a search experience that returns direct answers instead of links, but rapid model releases, high query volume, and the need for low latency and high throughput made… 20% improved over the base model on search Llama · Mistral · NVIDIA NeMo