SUCCESS STORY — FULLI
Fulli — More reliable AI agents, anchored in the Confluence documentation
To make its AI agents' answers reliable, Fulli asked KPC to anchor them in its Confluence documentation: vector indexing in Data 360 (formerly Data Cloud) and retrieval-augmented generation (RAG), with sources cited.
The impact in numbers
3
Stages: index, search, answer with the sources
The impact in numbers
3
Stages: index, search, answer with the sources
The impact in numbers
3
Stages: index, search, answer with the sources
RESULTS
The benefits observed for Fulli
Fewer hallucinations, answers you can check
Answers anchored in the company's real documentation rather than invented by the LLM, and checkable thanks to the cited Confluence sources.
The benefits observed for Fulli
Users trust it more
Greater day-to-day adoption of the AI agents, with fewer manual checks needed.
The benefits observed for Fulli
Continuous updating, with no retraining
Any change in Confluence is automatically reflected in the answers, which makes maintenance simpler.
RESULTS
The benefits observed for Fulli
Fewer hallucinations, answers you can check
Answers anchored in the company's real documentation rather than invented by the LLM, and checkable thanks to the cited Confluence sources.
The benefits observed for Fulli
Users trust it more
Greater day-to-day adoption of the AI agents, with fewer manual checks needed.
The benefits observed for Fulli
Continuous updating, with no retraining
Any change in Confluence is automatically reflected in the answers, which makes maintenance simpler.
RESULTS
The benefits observed for Fulli
Fewer hallucinations, answers you can check
Answers anchored in the company's real documentation rather than invented by the LLM, and checkable thanks to the cited Confluence sources.
The benefits observed for Fulli
Users trust it more
Greater day-to-day adoption of the AI agents, with fewer manual checks needed.
The benefits observed for Fulli
Continuous updating, with no retraining
Any change in Confluence is automatically reflected in the answers, which makes maintenance simpler.
Background & core challenges
AI agents whose answers were sometimes unreliable, and trust to win back
Fulli's AI agents were producing answers that were sometimes unreliable. The objective: make their answers reliable by anchoring them in the company's Confluence documentation.
Reduce hallucinations, the answers invented by the LLM
Make reliable answers that were imprecise, inconsistent and poorly suited to the business
Trace the source of every answer
Restore trust in the AI agents
Background & core challenges
AI agents whose answers were sometimes unreliable, and trust to win back
Fulli's AI agents were producing answers that were sometimes unreliable. The objective: make their answers reliable by anchoring them in the company's Confluence documentation.
Reduce hallucinations, the answers invented by the LLM
Make reliable answers that were imprecise, inconsistent and poorly suited to the business
Trace the source of every answer
Restore trust in the AI agents
Background & core challenges
AI agents whose answers were sometimes unreliable, and trust to win back
Fulli's AI agents were producing answers that were sometimes unreliable. The objective: make their answers reliable by anchoring them in the company's Confluence documentation.
Reduce hallucinations, the answers invented by the LLM
Make reliable answers that were imprecise, inconsistent and poorly suited to the business
Trace the source of every answer
Restore trust in the AI agents
Our approach
Index, search, answer with the sources
Index Confluence
Ingestion and vector indexing
Synchronizing the Confluence articles and pages through the Data Cloud connector, then vector indexing: Data Cloud splits the pages into "chunks".
Search
Retrieval: the most relevant passages
For every question, the agent runs a vector search across the index to find the most relevant Confluence passages.
Answer with the sources
Augmented generation
The LLM generates an answer anchored in Confluence's real content, with sources cited.
Our approach
Index, search, answer with the sources
Index Confluence
Ingestion and vector indexing
Synchronizing the Confluence articles and pages through the Data Cloud connector, then vector indexing: Data Cloud splits the pages into "chunks".
Search
Retrieval: the most relevant passages
For every question, the agent runs a vector search across the index to find the most relevant Confluence passages.
Answer with the sources
Augmented generation
The LLM generates an answer anchored in Confluence's real content, with sources cited.
Our approach
Index, search, answer with the sources
Index Confluence
Ingestion and vector indexing
Synchronizing the Confluence articles and pages through the Data Cloud connector, then vector indexing: Data Cloud splits the pages into "chunks".
Search
Retrieval: the most relevant passages
For every question, the agent runs a vector search across the index to find the most relevant Confluence passages.
Answer with the sources
Augmented generation
The LLM generates an answer anchored in Confluence's real content, with sources cited.
Our clients
Our clients
They trust us.
They trust us.
Equipment rental
Professional Equipment Rental Company | Industrialized tender response 100% Snowflake
Gen AI
Snowflake
→
Transport & Logistics
Global shipping company — Data Science: from R&D to industrialization
AI
Data
→
Industry
Aqualung Group — Transforming the data platform
Snowflake
Data Platform
→
Business services
APAVE — Cross-office capacity planning
Pigment
EPM
→
Organic food distribution
Biocoop — Steering the forecasting processes with SAP BPC
SAP
Retail & Distribution
→
Heating, plumbing and renewable energy distribution
Richardson — Opening up access to data by moving to the cloud
Qlik
Snowflake
→
Equipment rental
Professional Equipment Rental Company | Industrialized tender response 100% Snowflake
Gen AI
Snowflake
→
Transport & Logistics
Global shipping company — Data Science: from R&D to industrialization
AI
Data
→
Industry
Aqualung Group — Transforming the data platform
Snowflake
Data Platform
→
Business services
APAVE — Cross-office capacity planning
Pigment
EPM
→
Organic food distribution
Biocoop — Steering the forecasting processes with SAP BPC
SAP
Retail & Distribution
→
Heating, plumbing and renewable energy distribution
Richardson — Opening up access to data by moving to the cloud
Qlik
Snowflake
→
Equipment rental
Professional Equipment Rental Company | Industrialized tender response 100% Snowflake
Gen AI
Snowflake
→
Transport & Logistics
Global shipping company — Data Science: from R&D to industrialization
AI
Data
→
Industry
Aqualung Group — Transforming the data platform
Snowflake
Data Platform
→
Business services
APAVE — Cross-office capacity planning
Pigment
EPM
→
Organic food distribution
Biocoop — Steering the forecasting processes with SAP BPC
SAP
Retail & Distribution
→
Heating, plumbing and renewable energy distribution
Richardson — Opening up access to data by moving to the cloud
Qlik
Snowflake
→
AN AI AGENT PROJECT IN MIND?
AI agents that answer with your own sources?
KPC anchors your Agentforce agents in your company documentation through RAG on Data 360, for answers that are reliable and checkable.
AN AI AGENT PROJECT IN MIND?
AI agents that answer with your own sources?
KPC anchors your Agentforce agents in your company documentation through RAG on Data 360, for answers that are reliable and checkable.
AN AI AGENT PROJECT IN MIND?
AI agents that answer with your own sources?
KPC anchors your Agentforce agents in your company documentation through RAG on Data 360, for answers that are reliable and checkable.