{"slug":"franklin-templeton-scales-investment-insight-with-databricks-agent-bricks","url":"https://findausecase.com/use-cases/franklin-templeton-scales-investment-insight-with-databricks-agent-bricks","title":"Franklin Templeton scales investment insight with Databricks Agent Bricks","description":"Franklin Templeton built the internal SIGNALS application on Databricks Agent Bricks, combining proprietary fund scoring models with unstructured documents like prospectuses and fact sheets in Unity Catalog to generate AI-powered portfolio commentary. Analysts save 2+ hours per week, field teams surfaced $15 million in new product opportunities during early rollout, and AI-powered commentary coverage expanded from 200 products to hundreds.","company":"Franklin Templeton","industry":"Financial Services","aiCapabilities":["Generative AI","Document Intelligence"],"technology":["Databricks Agent Bricks","Unity Catalog","Databricks lakehouse"],"deployment":"Public Cloud","problemStatement":"The expanding universe of mutual funds, ETFs, digital assets and alternative investments made it challenging to deliver timely, customized insights. Legacy processes limited capacity for analysts -- a team of just seven who produced manually authored, highly researched notes -- and could support only 200 products with analyst-generated commentary, leaving many opportunities on the table. Standard dashboards and shared libraries couldn't keep pace, and early experiments with foundation models fell short because they generated text but lacked grounding in the firm's data, making them unusable in a regulated production environment.","solutionApproach":"Franklin Templeton built the internal SIGNALS application on Databricks Agent Bricks, combining proprietary fund scoring models with unstructured documents like prospectuses, fact sheets and portfolio manager commentary in Unity Catalog and the Databricks lakehouse architecture to generate AI-powered portfolio commentary. The team purpose-built several agents with Agent Bricks, including a custom LLM agent trained on 200 analyst-authored notes and two extraction agents that process PDF fund fact sheets and commentary into structured, machine-readable formats, with Agent Bricks' integrated evaluation tools used to tune prompts so AI commentary meets length, clarity and compliance requirements.","businessValue":"Analysts save 2+ hours per week, reclaiming more than 15 collective hours each week. Field teams surfaced $15 million in new product opportunities during the early rollout, and AI-powered commentary coverage expanded from 200 products to hundreds of funds and ETFs. SIGNALS was released to more than 300 salespeople in July.","evidence":{"band":"high"},"sourceUrl":"https://www.databricks.com/customers/franklin-templeton/agent-bricks","dates":{"publishedAt":"2026-08-16T00:04:35.298Z","publishedAtSource":"ledger","updatedAt":"2026-08-18T10:29:37.648Z"},"license":"Open for reading and citing with a link to https://findausecase.com/use-cases/franklin-templeton-scales-investment-insight-with-databricks-agent-bricks. Bulk republication requires permission."}