Domo

A natural-language interface that lets business users ask questions of their data and receive instant visualizations and narrative summaries. AI Chat is grounded in the user's governed Domo datasets, so answers respect existing permissions and data definitions rather than hallucinating off generic LLM knowledge. It supports follow-up questions, drill-downs, and explanations of how each answer was calculated.

A no-code/low-code framework for building custom AI agents that combine LLM reasoning with Domo data, workflows, and write-back actions. Users can define an agent's goal, give it access to specific datasets and tools, and deploy it to Slack, Teams, email, or directly into Domo dashboards. Agent Catalyst supports any approved LLM, so you can route different agents to different models based on cost or capability.

Domo.AI ships with prebuilt forecasting models for time-series prediction (revenue, demand, churn) that can be applied to any dataset without data-science expertise. The AI SQL Assistant generates and explains SQL for Magic ETL transformations, helping analysts move from natural-language intent to production-ready queries. Together they collapse the gap between BI users and ML/SQL practitioners.

FileSets allow Domo to ingest unstructured documents — PDFs, contracts, transcripts, emails — and make them queryable alongside structured data. Combined with AI Chat, this means a user can ask 'summarize all renewal risks across our Q4 customer call transcripts' and get a grounded answer. This brings RAG-style capabilities into a governed BI environment.

A central control plane for managing which LLMs and ML models are available, who can use them, and on which datasets. Admins can plug in OpenAI, Anthropic, Google, Amazon Bedrock, Hugging Face, or private models, set usage policies, and audit every AI interaction. This is a core requirement for regulated industries and is more mature in Domo than in most competing BI tools.