Conversation context
Multi-turn conversations and multimodal input
ERP AI KNOWLEDGE ASSISTANT
Business knowledge.
Inside your ERP.
An ERP-integrated assistant that makes onboarding, process guidance and organisational knowledge accessible within the user’s everyday workflow.
Customer production deployment · Anonymised case study
The challenge
People need reliable access to organisational knowledge while working in the ERP. Switching between applications and searching separate documents adds friction to onboarding and day-to-day process guidance.
HOW THE WORKFLOW CONNECTS
From request to useful response.
START INSIDE THE ERPERP user / conversation
A question in the context of everyday work
COORDINATION
ERP AI Knowledge Assistant
Conversation context · Knowledge retrieval · ERP integration
Knowledge retrieval
File search and vector stores for relevant organisational information
OpenAI processing
Responses API processing informed by the retrieved context
Validation & citations
Response validation and citation handling before delivery
BACK TO THE USERContextual answers inside the ERP
Support for onboarding, process guidance and organisational knowledge.
Simplified architecture illustrating how the ERP conversation connects to knowledge retrieval, AI processing and response handling.
FROM QUESTION TO GUIDANCE
Bring the knowledge closer to the work.
Ask within the existing workflow
The user starts a conversation inside the ERP. Instead of moving to a separate application, they can ask for onboarding help, process guidance or information from organisational knowledge.
Connect the question to relevant knowledge
Conversation context and knowledge retrieval work together to inform the response. File search and vector stores help locate relevant material rather than relying only on a model’s general knowledge.
Process the request with OpenAI
The integration uses the OpenAI Responses API as part of the assistant workflow. Multi-turn context supports follow-up questions, while the implementation also includes multimodal input capabilities.
Return a response with supporting context
Response validation and citation handling sit between AI processing and the ERP response. Citations help users understand the information behind an answer; they do not replace checking important decisions against the original source.
IMPLEMENTATION
What we built.
An assistant in the ERP
Bring a conversational interface into the existing ERP workflow, with support for multi-turn context.
Retrieval from knowledge
Use file search and vector stores to retrieve relevant information for the user’s question.
Responses with context
Connect OpenAI processing to response validation and citation handling before returning an answer in the ERP.
What this demonstrates
A delivered customer integration connecting ERP, knowledge retrieval and AI processing. The workflow illustrates how existing business software can provide AI assistance without requiring users to move to a separate application.
Explore Odoo AI integration services ↗APPLYING THIS TO YOUR BUSINESS
Your knowledge.
Your ERP workflow.
Start with the questions your team needs answered most often.
For a similar implementation, we would map the intended users, knowledge sources and ERP touchpoints first. Onboarding instructions, process documentation and internal guidance are practical areas to explore.
The project scope should define who may access which information, how knowledge is maintained, and how response quality will be evaluated. Those decisions shape the integration and rollout for your business.
Discuss an ERP knowledge assistantYOUR NEXT PROJECT