Knowledge graphs fuel AI innovation for financial services at Raymond James Financial
SiliconANGLE Mark Albertson
Raymond James is using knowledge graphs to make its AI agents safer and more useful. It wants answers grounded in trusted data, not dashboard guesswork.
Based on reporting by SiliconANGLE, Mark Albertson — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
Raymond James Financial is leaning on knowledge graphs as the quiet layer that makes its AI efforts less risky and more useful. At GraphSummit, Jeff Griffith, the firm’s vice president of AI engineering and data science, said graph technology helps the company connect data across business domains and build tools that do more than just repeat what already sits in front of users.
The appeal is pretty simple: finance needs accuracy, and AI needs context. As a Neo4j customer, Raymond James uses its graph to show relationships, dependencies and business meaning in a form that agents can reason over. Naveen Mahankali, the company’s vice president of technology, said the firm compares agent output against its “golden sources” and runs evaluations in Neo4j before anything goes live.
That matters because the goal is not just better back-office plumbing. Raymond James wants financial advisors to ask for what they need directly, instead of clicking through screens or staring at fixed dashboards. Mahankali said that dynamic querying lets advisors do more analysis, including product analysis and account-balance work, by asking questions rather than hunting through applications.
The company is already pushing this into everyday work. In January, it launched Raimond, its proprietary digital AI operations agent, and Griffith said adoption has been strong. He also said the broader shift inside the firm is changing what desktop work looks like: fewer menial tasks, more management of agents that handle them. That tends to be the real AI story in finance anyway — not magic, just a lot of very careful software wrapped around very expensive trust.
My take — AI-written commentary, not fact-checked reporting
Finance is one of the few places where the boring answer is the right one. Knowledge graphs, golden sources, and evals before launch are not sexy, but they beat the usual AI theatre by a mile. The real innovation here is refusing to let the model freestyle with other people’s money.
Read more about this at: SiliconANGLE