Formula 1 reduces data source onboarding time from weeks to minutes using agentic AI on AWS
F1’s MarTech platform integrated agentic AI via Amazon Bedrock AgentCore to automate schema evolution and data pipeline observability, cutting new source onboarding from up to 8 weeks to about 40 minutes.
1 source · cross-referenced
- Formula 1 partnered with AWS to build an agentic AI system on Amazon Bedrock AgentCore to automate data operations for its MarTech platform.
- The system reduced new data source onboarding time from up to 8 weeks to about 40 minutes.
- Agentic AI automated schema evolution and provided end-to-end observability across F1’s fan-engagement data estate.
- F1’s Customer 360 platform powers personalization and commercial strategy for over 800 million global fans.
Formula 1® (F1) integrated agentic AI on AWS to automate data operations for its Customer 360 MarTech platform, which supports personalization and commercial strategy for more than 800 million global fans. The system was built using Amazon Bedrock AgentCore and aimed to reduce the time required to onboard new data sources, which previously took between six and eight weeks of manual engineering per source.
According to Chris Roberts, Director of IT at Formula 1, F1’s MarTech platform faced an 18‑month backlog to integrate 12 new data sources, with data generation outpacing the engineering team’s ability to wire it up. F1’s Head of Data Operations, Matt Kemp, described manual ingestion as time‑consuming, error‑prone, and a source of data integrity issues, prompting the search for a repeatable, robust, and reliable solution.
The agentic AI system automated schema evolution and provided end‑to‑end observability across F1’s fan‑engagement data estate. By leveraging Amazon Bedrock AgentCore, F1 reduced the onboarding time for new data sources from up to eight weeks to about 40 minutes, enabling faster integration of marketing and fan‑engagement data sources.
The implementation targeted F1’s Customer 360 platform, which aggregates interactions across digital platforms, F1 TV, social media, ticketing, and merchandise to power segmentation and commercial decisions that must move at the speed of the racing calendar.
- Aug 3, 2026 · Microsoft Research
Microsoft Research releases Orchard, an open framework for training and evaluating agentic AI systems
Trust84 - Aug 3, 2026 · arXiv cs.AI
Paper proposes layered architecture for Agentic AI using OpenClaw and Ollama
Trust79 - Aug 1, 2026 · Microsoft Research
Microsoft Research releases Echoverse, a framework for training computer-use agents in deep, evolving environments
Trust79