LLAMA 3.1 70B INSTRUCT
Sevilla Football Club
Making talent scouting faster and easier with Llama
At a glance
Industry: Professional sports
Use case: Prompt enrichment, semantic search and summarization
Goal: Search and summarize unstructured scouting reports quickly and easily
Llama versions:Llama 3.1 70B Instruct
Deployment: IBM watsonx running on IBM Cloud
From hours to seconds
Scouting distilled
Tech leadership
New revenue
*All results are self-reported and not identifiably repeatable. Generally expected individual results will differ.
Driving excellence from the pitch to the back office
Seven-time Europa League champions Sevilla FC uses data to stay competitive on and off the pitch. The team fields a roster of in-house AI tools, including Scout Advisor, AiFootball, AiFans and AiTicketing, fueling an innovative, data-driven culture that helps keep Sevilla FC at the top of its game.
THEIR GOAL
Transform days of scouting research into quick conversations
Sevilla FC developed intelligent tools for match analysis, player performance and fan marketing. The tools were excellent with quantitative data but struggled with unstructured information like scouting reports. Scouting research and analysis had to be done by hand in a process that took days.
The research is worth the effort because expert scouts capture characteristics like tenacity, endurance, and attitude — traits as unique as each player. With a database of more than 300,000 scouting reports, hand analysis was becoming untenable. The team set an ambitious goal: use AI to make searching scouting reports as easy as describing an ideal player in your own words.
THEIR SOLUTION
A smart scouting agent that generates reports on demand
The Sevilla FC Data Department used Llama 3.1 70B Instruct and the IBM watsonx platform to create Scout Advisor, a generative AI application that provides conversational search tools, curated results and player summaries in an intuitive, user-friendly platform.
Llama powers an intelligent tool that synthesizes player information from more than 300,000 scouting reports.
THEIR APPROACH
Teaching Llama the language of futból — en Español, tambien
Futból (also known as soccer) is filled with idiomatic phrases and insider jargon. Talent scouts add layers of professional lingo and individual idiosyncrasies, creating a lexicon that general-purpose models can’t penetrate.
For example, asking the average AI assistant to “show me talented wings” might provide pilot biographies, bird descriptions, or chicken recipes. Ask Scout Advisor to “show me talented wings” and Llama will enrich the prompt with soccer-specific context like “a talented wing takes on defenders with dribbling, creating space and penetrating the opposition.”
The enriched prompt feeds a similarity search that pulls relevant scouting report data and injects it — with the embedded prompt — into a Llama-powered engine for answer generation. The result: a list of talented soccer wings with a comprehensive summary of each player’s scouting reports — all in Spanish.
Sevilla FC’s Scout Advisor uses Llama 3.1 70B Instruct running on IBM watsonx in a retrieval augmented generation (RAG) pipeline.
THEIR SUCCESS
Faster player evaluations, new revenue streams
By putting nuanced scouting information at recruiters’ fingertips, Llama-based Scout Advisor has helped reinvent how Sevilla FC tracks, analyzes and evaluates players. In addition to improving performance on the pitch, the Scout Advisor has made Sevilla FC a global leader in sports AI. Their new reputation is attracting players and creating business opportunities with other teams who want to use AI in their operations.
• From hours to seconds
- • Reduced evaluation time for multiple scouting reports on an individual player from hours to seconds
• Scouting distilled
- •Enabled identification of players based on a large dataset of human expert opinions
• Tech leadership
- •Positioned Sevilla FC as a global AI leader in the sports industry
• New revenue
- •Created consulting opportunities and new revenue streams
Scout Advisor uses Llama 3.1 70B Instruct’s advanced natural language processing to bridge the gap between qualitative human insights and quantitative data analysis. This fusion enhances the efficiency and effectiveness of our scouting operations, ensuring that our recruitment strategies are both data-driven and deeply informed by human expertise.
"Scout Advisor uses Llama 3.1 70B Instruct’s advanced natural language processing to bridge the gap between qualitative human insights and quantitative data analysis. This fusion enhances the efficiency and effectiveness of our scouting operations, ensuring that our recruitment strategies are both data-driven and deeply informed by human expertise."
Elías Zamora, Chief Data Officer (CDO), Sevilla Fútbol Club
Models used
Create generative AI applications for business with open-source large language models that bring unmatched control, customization and flexibility.
Llama 3.1 70B Instruct
Stay up-to-date
Our latest updates delivered to your inbox
Subscribe to our newsletter to keep up with the latest Llama updates, releases and more.