Back office ops · Production

Intuit builds a GenAI-powered dual-loop pipeline to transform document management and knowledge discovery

The problem

Intuit's technical documentation suffered from inconsistent quality, difficulty determining whether information was current, poor structure for information retrieval, and content not written with target audiences in mind — making it hard for engineers to find the right information at the right time.

Workflow diagram · grounded in source
1
Document quality scoring
validation
“This tool looks at a document's structure, completeness and ease of comprehension. It compares these elements against a custom rubric to come up with a score for the document. Any element that does not meet the required threshold moves t…”
2
Content restructuring
ai_action
“the improvement plugin is a skilled writer that restructures and enhances the content. The tool restructures the content with an eye toward making it more coherent and comprehensive.”
3
Style guide enforcement
ai_action
“This tool modifies the voice and style of a document to ensure consistency across the entire knowledge base.”
4
Discoverability optimization
ai_action
“This plugin modifies content to ensure the outer system can find it when it needs it. It accomplishes this by adding semantic context and linking it to relevant user queries.”
5
RAG-based content augmentation
ai_action
“This plugin uses retrieval-augmented generation to pull in new, relevant information from various knowledge sources that apply to the document in question. It uses this information to revise the documentation as necessary to keep it up t…”
6
Vector embedding creation
ai_action
“This tool creates vector representations of documents that serve as the basis for sophisticated similarity searches and content clustering.”
7
Semantic similarity search
ai_action
“This plugin uses semantic similarity to scan prospective areas of content and find the most relevant chunks to respond to user queries.”
8
Answer synthesis
output
“This plugin brings the pieces together by synthesizing the information provided by the other two plugins into comprehensive, accurate answers to user queries.”
9
Feedback-driven improvement
feedback_loop
“When the system is unable to answer a query successfully, it updates the Search Plugin to improve results for future similar searches and updates the base documents with any missing information.”
Reported outcome

The GenAI pipeline improves documentation quality and discoverability, reduces time engineers spend searching for information, and provides knowledge workers with context-aware comprehensive answers to their queries.

Reported metrics
Time spent seeking informationless time spent seeking out information
Knowledge worker query experiencebetter user experience
Reported stack
Large language models (LLMs)vector stores
Source
https://medium.com/intuit-engineering/revolutionizing-knowledge-discovery-with-genai-to-transform-document-management-0cdf4385c11c
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

The GenAI pipeline improves documentation quality and discoverability, reduces time engineers spend searching for information, and provides knowledge workers with context-aware comprehensive answers to their queries.

What tools did this team use?

Large language models (LLMs), vector stores.

What results were reported?

Time spent seeking information: less time spent seeking out information; Knowledge worker query experience: better user experience (source-reported, not independently verified).

How is this back office ops AI workflow structured?

Document quality scoring → Content restructuring → Style guide enforcement → Discoverability optimization → RAG-based content augmentation → Vector embedding creation → Semantic similarity search → Answer synthesis → Feedback-driven improvement.

Related back office ops cases