Bundesliga builds Captain, an agentic AI companion for fans, on Amazon Bedrock
Bundesliga Media built Captain, an AI companion embedded in the official Bundesliga app that answers fan questions via text-to-SQL over match data and runs research agents that surface personalised insights as events happen. A model-routing layer sends about 45% of questions to Amazon Nova Pro and the rest to Claude Sonnet 4; the AWS blog reports a 35% reduction in chat assistant service costs with the same answer accuracy.
Overview
Bundesliga Media built Captain, an AI companion embedded in the official Bundesliga app that answers fan questions via text-to-SQL over match data and runs research agents that surface personalised insights as events happen. A model-routing layer sends about 45% of questions to Amazon Nova Pro and the rest to Claude Sonnet 4; the AWS blog reports a 35% reduction in chat assistant service costs with the same answer accuracy.
This entry has 11 published fields tied to exact passages in an immutable source capture.
Inspect the highlighted sourceThe challenge
Bundesliga wanted to give fans a single conversational interface for live stats, historical trivia, tactical analysis and video highlights, and to surface insights that fans would never find on their own or didn't know to ask for.
The solution
Bundesliga Media built Captain, an agentic AI companion embedded in the official Bundesliga app. A chat service uses a text-to-SQL workflow in which an LLM converts a fan's question into a database query that Amazon Athena runs against statistical data stored in Amazon S3 Tables, and a second LLM call turns the results into a natural-language answer. A model routing layer classifies each question and sends straightforward ones to Amazon Nova Pro and medium and complex ones to Claude Sonnet 4 in Amazon Bedrock. A research service built on the Strands Agents SDK with Amazon Bedrock triggers agents on events such as goals to surface personalized insights, and Amazon Bedrock Guardrails screen inputs and check that outputs are grounded.
Reported business value
Roughly 45% of fan questions qualify for the more economical Nova Pro path, and this dynamic approach delivered a 35% reduction in chat assistant service costs while maintaining the same answer accuracy.
Sources
Open any source and check the claim yourself — that is the point of the register.
This record was researched and written with AI assistance, and its claims were checked against the sources above. (EU AI Act art. 50 transparency notice.)
Other media & entertainment entries in the register.
How Business Insider's AI-based paywall strategy increased conversions by 75%
Business Insider replaced its editorially-driven freemium paywall with a machine-learning-based smart paywall that decides which content to paywall for each user based on prior reading habits, referral platform, and each content genre's propensity to convert. In testing from December to April, total conversions increased more than 75% versus the control, with 60% of new conversions coming from stories that would never have been paywalled before. Applying AI to the registration wall as well increased registrations by 300%.
Leveraging hyper-personalization: Enriching the player journey through data and automation
Games24x7, an Indian online gaming platform with over 100 million players, partnered with Databricks to unify data operations and automate data pipelines supporting its machine-learning-driven hyper-personalization and fraud detection systems. The move deployed fraud prediction models 16% faster and reduced operational costs for BI workloads by 20%, while a marketing optimization platform built on the Databricks Data + AI Platform tracks and scales campaign performance.
Delivering AI-powered entertainment that captivates users for hours
Scatter Lab used Databricks' flexible GPU clusters to build a proprietary Korean-language large language model in three months on a limited budget, powering its AI companion chatbot platform Zeta, which reached 1.5 million users within nine months with over 12 hours of average weekly engagement.
Süddeutsche Zeitung: Lokalinformationen per WhatsApp
Süddeutsche Zeitung (SZ) launched WhatsApp channels for districts around Munich, offering free local news, event information and regional updates. Content comes from SZ's local newsrooms and is supplemented by automatically generated information produced in cooperation with Berlin media startup Beat Squares, which uses AI to process data from publicly accessible sources. All contributions are editorially reviewed before publication. More than 4,000 users subscribed within the first three days.
Was this helpful?
Your feedback helps us improve our use case database
