AWS re:Invent 2025 - Formula 1 Extends Fandom with AI, Data, and AWS Media Services (IND3326)

Source: Dev.to

Introduction

AWS and Formula One demonstrate how they deliver F1 TV Premium, a multi‑screen streaming platform that processes over 5 million data points per race weekend. The presentation highlights two major innovations:

• A root‑cause analysis system built with Amazon Bedrock multi‑agent orchestration that cut ticket‑resolution time by 95 % (from 45 minutes to 4 minutes).
• The technical architecture behind F1 TV Premium’s synchronized multi‑view experience, using AWS Elemental MediaLive with epoch locking, HEVC tiled encoding, and timecode alignment across 24 video feeds to enable frame‑accurate synchronization of multiple camera angles in UHD HDR at 50 fps.

The platform launched successfully at the 2025 season opener in Australia.

Speakers

• Nick Morgan – Solutions Architect, AWS
• Jamie Mullan – Specialist Solutions Architect, Edge Services, AWS
• Dave King – Head of Digital Technology, Formula One

Formula One & AWS Partnership Overview

• AWS began working with Formula One in 2018 and became a global sponsor for the 2019 season.
• The partnership was renewed in November 2022 and now sits at the top tier of Formula One collaborations.
• AWS also partners with other sports leagues (NFL, PGA, Bundesliga), but F1’s data‑centric nature makes the relationship unique.

Key Collaboration Pillars

1. Transformation of data into racing intelligence
2. Fan‑experience enhancement
3. Technical transformation through joint technology funds

Notable Milestones

Data‑Intensive Nature of F1

• Each F1 car generates over 1 million data points per second.
• Across a typical race weekend, more than 5 million data points are processed from all cars – roughly equivalent to streaming 1 500 HD movies simultaneously.

Data Capture at the Track

• Induction loops are placed roughly every 200 m around the circuit, forming sectors for timing data.
• Radio towers receive live telemetry and onboard video from the cars.

Data Transport to AWS

1. Event Technical Center (ETC) – a portable mini‑data‑center located at the circuit, connected via high‑speed optical links. It houses a video production gallery.
2. Media and Technology Centre (MNTC) in Biggin Hill, Kent – linked to the ETC through fiber optics; contains a full broadcast suite, three studios, commentator spaces, and IT/engineering departments.
3. AWS Direct Connect – the MNTC uses Direct Connect to send a subset of telemetry data and finished video feeds (world feed, onboard cameras, pre‑canned graphics) to AWS.

AWS Architecture for F1

Data Lake & Storage

• Amazon S3 stores:
  • Trained ML models ready for inference
  • The Customer 360 data lake – a single source of truth for fan profiles and interaction data from F1.com, F1 TV, etc.
  • Video archive – over 7 PB of historical footage

Machine Learning & Insights

• Amazon SageMaker – inference engine for F1 Insights; models are retrained between race weekends.
• Inference pipeline:
  1. Data arrives via Amazon API Gateway.
  2. Processing runs in AWS Lambda functions.
  3. Results are stored in Amazon DynamoDB (NoSQL key‑value store).
  4. DynamoDB also holds model‑training artifacts for continuous improvement.
• Amazon CloudWatch – collects operational logs; its metrics feed back into the insight system to monitor system health.

The insight data is then delivered to video production teams for use during broadcasts.

F1 TV Premium Multi‑Screen Platform

Media Processing

• AWS Elemental MediaLive with epoch locking ensures all video feeds share a common time reference.
• HEVC tiled encoding enables efficient delivery of multiple camera angles.
• Timecode alignment across 24 video feeds provides frame‑accurate synchronization.

Fan Experience

• Viewers can watch multiple camera angles simultaneously in UHD HDR at 50 fps.
• The platform supports a synchronized multi‑view experience across devices (TV, mobile, web).

Root‑Cause Analysis System

• Built on Amazon Bedrock with multi‑agent orchestration.
• Automates detection and diagnosis of operational issues, reducing ticket‑resolution time from 45 minutes to 4 minutes (a 95 % improvement).

Impact & Results

These improvements enhance both operational efficiency for Formula One and the viewing experience for fans worldwide.

Conclusion

The collaboration between AWS and Formula One showcases how advanced cloud services, AI/ML, and media processing can transform a data‑heavy sport into a richer, more interactive fan experience. From real‑time telemetry insights to a synchronized multi‑screen streaming platform, the partnership continues to push the boundaries of what’s possible in live sports broadcasting.