Each colour maps a published value to the exact source passage used to support it. Only bounded excerpts are public; administrators can inspect the complete captured source.
Title
Deutsche Bank builds DB Lumina, a RAG research agent that saves analysts up to two hours per report
…time savings, saving 30 to 45 minutes on preparing earnings note templates and up to two hours when writing research reports and roadshow updates. Increased analysis depth: One analyst increased the analysis in an earnings re…
Description
Deutsche Bank Research built DB Lumina, an AI research agent on Google Cloud (GKE, Vertex AI Gemini models, Cloud SQL pgvector, Discovery Engine RAG) that lets analysts chat with documents, use standardized prompt templates, and ground answers in enterprise knowledge with inline citations. Since going live in September 2024, DB Lumina has reached about 5,000 users across Deutsche Bank Research, saving analysts 30-45 minutes preparing earnings note templates and up to two hours writing research reports and roadshow updates, with one analyst increasing report analysis depth by 50%.
…ronments to our production deployment, eventually going live in September 2024. Currently, DB Lumina is already in the hands of around 5,000 users across Deutsche Bank Research, specifically in divisions like Investment Bank Origination & Advisory and Fixe…
…s when writing research reports and roadshow updates. Increased analysis depth: One analyst increased the analysis in an earnings report by 50%, adding additions sections by region and activity, as well as a summary section…
Company
Deutsche Bank
…oogle Try Gemini Enterprise today The front door to AI in the workplace Try now At Deutsche Bank Research , the core mission of our analysts is delivering original, independent economic and financial analysis. However, creating research reports and notes relies heavily on a foundation of…
Industry
Financial Services
…oogle Try Gemini Enterprise today The front door to AI in the workplace Try now At Deutsche Bank Research , the core mission of our analysts is delivering original, independent economic and financial analysis. However, creating research reports and notes relies heavily on a foundation of…
Problem
Deutsche Bank Research analysts relied on painstaking manual work to create research reports: sifting through and gathering data from financial statements, regulatory filings and industry reports, then synthesizing vast amounts of information to build financial models and identify patterns, which limited the depth of analysis and range of topics analysts could cover.
…s is delivering original, independent economic and financial analysis. However, creating research reports and notes relies heavily on a foundation of painstaking manual work. Or at least that was the case until generative AI came along. Historically, an…
Solution
Deutsche Bank built DB Lumina with a gen AI-powered chat interface using Google's Gemini foundation models, prompt templates for consistent document processing, and a retrieval-augmented generation architecture grounding responses in enterprise knowledge sources with inline citations, built on GKE, Cloud SQL with pgvector, Cloud Storage, Dataflow, Vertex AI and the Discovery Engine API, with guardrailing techniques for compliant outputs.
…ompting techniques, which also provide citations for verification. 3. Knowledge DB Lumina integrates a Retrieval-Augmented Generation (RAG) architecture that grounds responses in enterprise knowledge sources, such as internal research, external unstructured data (such as SEC filings), a…
Business value
Since going live in September 2024, DB Lumina reached about 5,000 users across Deutsche Bank Research, saving analysts 30 to 45 minutes preparing earnings note templates and up to two hours writing research reports and roadshow updates, with one analyst increasing the analysis depth of an earnings report by 50% and improved editorial and grammatical accuracy in analyst notes.
…ts for Deutsche Bank: Time savings: Analysts reported significant time savings, saving 30 to 45 minutes on preparing earnings note templates and up to two hours when writing research reports and roadshow updates. Increased analysis depth: One analyst increased the analysis in an earnings re…
Technology
Google Kubernetes Engine, Vertex AI, Cloud SQL, Cloud Storage, Dataflow, Discovery Engine API, Cloud Natural Language APIs, Gemini
…ting. Built on Google Cloud, the architecture leverages the following services: Google Kubernetes Engine (GKE) for microservice orchestration Cloud SQL with the pgvector extension for vector support Cloud Storage for managing and storing unstructured data Dataflow for document ingestion and embedding Vertex AI for powering multimodal AI capabilities with Gemini Discovery Engine API to enable RAG Cloud Natural Language APIs for text and content moderation All of DB Lumina’s…
AI capabilities
Retrieval-Augmented Generation, Generative AI, Large Language Models
…analyst experience and reduce the reliance on manual processes and outsourcing. We created DB Lumina — an AI-powered research agent that helps automate data analysis, streamline workflows, and deliver more accur…
Use case type
Analytics augmentation
…nt information when creating research reports and notes. DB Lumina architecture DB Lumina was designed to enhance Deutsche Bank Research’s productivity by enabling document ingestion, content summarization, Q&A, and editing. Built on Google Cloud, the architecture leverages the following services: Googl…
Headline outcome
…time savings, saving 30 to 45 minutes on preparing earnings note templates and up to two hours when writing research reports and roadshow updates. Increased analysis depth: One analyst increased the analysis in an earnings re…
Deployment model
cloud
…tivity by enabling document ingestion, content summarization, Q&A, and editing. Built on Google Cloud, the architecture leverages the following services: Google Kubernetes Engine (GKE) for microservice orchestration Cloud SQL with t…
Deployment options
cloud
…tivity by enabling document ingestion, content summarization, Q&A, and editing. Built on Google Cloud, the architecture leverages the following services: Google Kubernetes Engine (GKE) for microservice orchestration Cloud SQL with t…
