Frequently Asked Questions¶
This section answers common questions about getting started with the Intelligence Hub and about how your data is handled. For the authoritative statement on data handling, always refer to the Terms of Service you accept before using the agents.
Getting Started¶
What is the Intelligence Hub, and which agent should I use?
The Intelligence Hub is a web application with three specialized agents.
Use Gurobot for expert Gurobi support—questions, code
snippets, and documentation. Use Explainer to diagnose
infeasibility, repair feasibility, and run sensitivity analysis on a model
you upload. Use Modeler to turn a plain-language business
problem into a complete gurobipy project with a specification,
implementation, and tests. See the Overview for how the agents
compare and what stage each is at.
How do I access it?
The Intelligence Hub is available from the Gurobi portal and directly at intelligence.gurobi.com. The first time you sign in, you must review and accept the Terms of Service before you can use the agents.
Do I need to install anything?
No—the agents run on the Gurobi Intelligence Hub backend and you access them through your browser, with nothing to install. If you want to use them from an IDE or AI coding assistant instead, you can run the MCP server locally, which requires Docker.
How does the agent know about the latest Gurobi behavior?
All agents are grounded in a Gurobi-curated knowledge base that is re-indexed daily and cite their sources, so answers
reflect current Gurobi guidance rather than generic web content. When writing
gurobipy code, they also consult a dedicated coding guide.
What files can I work with, and where do they go?
Each conversation has its own workspace where uploaded and generated files are stored. Agents can also run code in a sandboxed code execution environment to verify what they produce. See the individual agent pages for the file types each one accepts and generates.
Is there a usage limit?
Yes. Daily usage limits apply across all agents, and a daily usage indicator on the home page shows how much of your limit you have consumed. Access may be temporarily suspended if you exceed the limit, and usage resets daily.
How do I give feedback?
Every agent response has thumbs up / thumbs down controls. See Feedback for how that feedback is used.
Security and Privacy¶
The answers below summarize how the Intelligence Hub handles your data. The binding terms are in the Terms of Service.
Are my inputs or outputs—uploaded files, prompts, responses, or chat history—used to train, retrain, or fine-tune any LLM?
No. Your interactions, inputs, outputs, and model input files are never used to train, fine-tune, or otherwise improve any large language model. See the Data is Not Used for Training LLMs clause of the Terms of Service.
Are my inputs or outputs used for analytics or product improvement?
Yes, in a limited way. Gurobi reserves the right to process your interactions to validate, upgrade, maintain, and otherwise develop the service—this does not include training LLMs (see above). In practice, the main driver of product improvement is your feedback, which Gurobi’s optimization experts use to refine the knowledge base and the agents’ instructions. See the Service Improvement clause of the Terms of Service.
Is my data isolated to my account, or is any of it processed in a shared environment that other organizations could see?
Your inputs are processed in a dedicated environment with no shared state or filesystem with any other session or user. Only the portions of your input needed to fulfill a request are transmitted to a large language model running within Gurobi’s secure cloud infrastructure. See the Security points in the Terms of Service for the full description of isolated processing and limited exposure.
How is my data protected in transit and at rest?
Inputs are encrypted in transit with HTTPS and encrypted at rest using application-layer encryption with a unique key per file. See the Security points in the Terms of Service.
Can I delete my data?
Yes. You can delete an individual conversation, all chats for an agent, or all
chats across every agent at any time from the Settings page.
Your inputs are permanently deleted when you do so, when you delete your
account, or when you send a written deletion request to
operations@gurobi.com, subject to applicable legal retention requirements.
See the Deletion point in the Terms of Service.
Should I include personal or confidential data in my prompts?
No. Do not include data that identifies or could reasonably be linked to a natural person—names, addresses, social security numbers, payment card information, and the like. See the Personal data point in the Terms of Service.
How is the MCP server kept secure?
The local MCP server is sandboxed to the single root directory you mount, passes your credentials only as environment variables rather than storing them in a project config, and exposes its activity through standard Docker logs. See the Design Rationale in the MCP server page.
Is the agent’s output guaranteed to be correct?
No. The service uses generative AI and may produce inaccuracies. Independently verify any output before relying on it. See the Generative AI disclaimer in the Terms of Service.