ML & AI Engineer specialized in football (and other sports) AI & Analytics, with 6 years of hands-on Python and ML practice, and an MSc in Data Science from Universität Potsdam (Germany), in collaboration with Hasso Plattner Institut (HPI). At EY I single-handedly founded, designed, and successfully built a Sports AI & Analytics business line from scratch (with a special focus on football), under the direct supervision and sponsorship of a department-linked share-holding Partner, in parallel to my formal duties. He supported carrying out this initiative after I proactively proposed it, shortly after joining EY, pitching the potential of this idea and business opportunity to the entire department. I led a small team of 3 technical people, shipping end-to-end ML, Deep Learning, and Generative AI/Agentic AI solutions mainly on: Databricks, LangChain/LangGraph/LangSmith, and GCP. Collaborated cross-functionally with technical and non-technical client stakeholders. During my MSc. Data Science thesis I developed a cutting-edge, state-of-the-art research project under the direct supervision from Dr. Gabriel Anzer (Head of Football Data Analytics at RB Leipzig) - successfully demonstrated statistically that Heterogeneous Graph Transformers (HGTs) classify, estimate, and calibrate better football shots & their respective goal probability (i.e. xG = Expected Goals), compared to conventional xG models; this work was selected by Founder David Sumpter as 1 of 4 finalists at Twelve Football's “Pitch to the Pros #3”. Available for freelance and contract engagements as an AI Engineer, Agentic AI Engineer, GenAI Engineer, ML Engineer, DL Engineer or AI-focused Data Scientist.…

Pedro Satorre-Mulet

ML & AI Engineer specialized in football (and other sports) AI & Analytics, with 6 years of hands-on Python and ML practice, and an MSc in Data Science from Universität Potsdam (Germany), in collaboration with Hasso Plattner Institut (HPI). At EY I single-handedly founded, designed, and successfully built a Sports AI & Analytics business line from scratch (with a special focus on football), under the direct supervision and sponsorship of a department-linked share-holding Partner, in parallel to my formal duties. He supported carrying out this initiative after I proactively proposed it, shortly after joining EY, pitching the potential of this idea and business opportunity to the entire department. I led a small team of 3 technical people, shipping end-to-end ML, Deep Learning, and Generative AI/Agentic AI solutions mainly on: Databricks, LangChain/LangGraph/LangSmith, and GCP. Collaborated cross-functionally with technical and non-technical client stakeholders. During my MSc. Data Science thesis I developed a cutting-edge, state-of-the-art research project under the direct supervision from Dr. Gabriel Anzer (Head of Football Data Analytics at RB Leipzig) - successfully demonstrated statistically that Heterogeneous Graph Transformers (HGTs) classify, estimate, and calibrate better football shots & their respective goal probability (i.e. xG = Expected Goals), compared to conventional xG models; this work was selected by Founder David Sumpter as 1 of 4 finalists at Twelve Football's “Pitch to the Pros #3”. Available for freelance and contract engagements as an AI Engineer, Agentic AI Engineer, GenAI Engineer, ML Engineer, DL Engineer or AI-focused Data Scientist.…

Available to hire

ML & AI Engineer specialized in football (and other sports) AI & Analytics, with 6 years of hands-on Python and ML practice, and an MSc in Data Science from Universität Potsdam (Germany), in collaboration with Hasso Plattner Institut (HPI).

At EY I single-handedly founded, designed, and successfully built a Sports AI & Analytics business line from scratch (with a special focus on football), under the direct supervision and sponsorship of a department-linked share-holding Partner, in parallel to my formal duties. He supported carrying out this initiative after I proactively proposed it, shortly after joining EY, pitching the potential of this idea and business opportunity to the entire department. I led a small team of 3 technical people, shipping end-to-end ML, Deep Learning, and Generative AI/Agentic AI solutions mainly on: Databricks, LangChain/LangGraph/LangSmith, and GCP. Collaborated cross-functionally with technical and non-technical client stakeholders.

During my MSc. Data Science thesis I developed a cutting-edge, state-of-the-art research project under the direct supervision from Dr. Gabriel Anzer (Head of Football Data Analytics at RB Leipzig) - successfully demonstrated statistically that Heterogeneous Graph Transformers (HGTs) classify, estimate, and calibrate better football shots & their respective goal probability (i.e. xG = Expected Goals), compared to conventional xG models; this work was selected by Founder David Sumpter as 1 of 4 finalists at Twelve Football’s “Pitch to the Pros #3”.

Available for freelance and contract engagements as an AI Engineer, Agentic AI Engineer, GenAI Engineer, ML Engineer, DL Engineer or AI-focused Data Scientist.

See more

Skills

Experience Level

VS Code
Expert
Spark
Expert
PyTorch
Expert
Python
Expert
Data Science
Expert
Data Visualization
Expert
Power BI
Expert
ML
Expert
Keras
Expert
Large Language Model
Expert
AI Chatbot
Expert
Prompt Engineering
Expert
Google Cloud
Expert
Github
Expert
Azure
Expert
API
Intermediate
Jira
Intermediate
SQL
Intermediate
Tableau
Intermediate
AWS
Beginner
Amazon Web Services
Beginner
TensorFlow
Beginner
See more

Language

Spanish; Castilian
Fluent
Catalan; Valencian
Fluent
English
Fluent
Greek, Modern
Intermediate
German
Beginner
Italian
Beginner
Chinese
Beginner

Work Experience

Agentic AI Engineering – Football/Soccer CBs & CB-Pairings AI Analyst Agent at Twelve Football
February 24, 2026 - May 29, 2026
▪ I got selected onto Twelve Football’s 1st edition of the course (AI) “Context Engineering for Football” (a very limited cohort of 45 people worldwide was hand-picked by Founder David Sumpter himself + his comittee) - implemented a hands-on project as part of the course, to develop an AI Analyst agent; ▪ I implemented a conversational CBs & CB-Pairings AI Analyst agent: grounding an LLM in structured football event data through context engineering, RAG, and tool-use techniques. It is then delivered as a Streamlit chat front-end app, answering natural-language football scouting and analyst questions with professional-analyst-grade player comparisons & pairing recommendations. ▪ AI Analyst Agents for other positions on the pitch are on their way...
Technology Consultant – AI, Data Science & Analytics -- (ML/AI Engineer) at Ernst & Young (EY)
October 14, 2024 - November 20, 2025
▪ Designed and built a new Sports AI & Analytics business line from scratch – single-handedly and as a parallel initiative alongside daily delivery – after proactively pitching the idea & business opportunity to the entire department, securing sponsorship and direct supervision from a department-linked share-holding Partner. ▪ Delivered end-to-end Machine Learning (ML) and Deep Learning (DL) solutions for clients: data preparation and feature engineering, model training and evaluation, experiment tracking, MLOps, deployment and monitoring. ▪ Developed Generative (GenAI) & Agentic AI solutions for clients: LLM-based knowledge assistants, retrieval-augmented (RAGs), and tool-using agents. ▪ Led a team of three technical specialists: task planning, technical guidance, code review, and delivery quality. ▪ Built decision-ready dashboards and visual analytics: translating model output into recommendations for business stakeholders. ▪ Ran cross-functional collaboration with technical and non-technical client teams, from requirements workshops through to hand-over.
Machine Learning & Deep Learning Researcher – Football AI & Data Analytics at RB Leipzig (German 1. Bundesliga Football Club)
June 1, 2023 - January 31, 2024
▪ State-of-the-art and cutting-edge MSc. Data Science thesis’ research project – modelling football shots as heterogeneous graphs of all players, the ball, as well as the instantaneous spatial & physical context, and then applying Heterogeneous Graph Transformers (HGTs) to classify shot outcomes and estimate their respective goal probability (i.e. xG = Expected Goals); ▪ Successfully demonstrated statistically that HGTs significantly improve the shot outcome classification, and their respective goal-probability estimation & calibration over traditional xG models. This thesis was directly supervised under Dr. Gabriel Anzer (Head of Football Data Analytics at RB Leipzig). ▪ In addition, this project was selected by Founder David Sumpter as 1 of 4 finalists for Twelve Football’s “Pitch to the Pros #3”.

Education

MSc. Data Science at Universität Potsdam (in cooperation with Hasso Plattner Institute - HPI)
October 26, 2020 - March 12, 2024
▪ Delivered in cooperation with the Hasso Plattner Institute (HPI) – Top-tier #1 German Institution for CS & IT. ▪ Main Courses : Advanced Machine Learning · Deep Learning · Data Science · Applied ML in Digital Health · Statistical Data Analysis · Business Data Analytics & BI · Bayesian Inference & Data Assimilation · Data- & Knowledge-Base Systems · Biostatistics & Epidemiological Data Analysis ▪ Thesis : “Shot Classification & Goal Probability Estimation Using Graph Neural Networks” (specifically, Heterogeneous Graph Transformers – HGTs) – directly supervised by Dr. Gabriel Anzer : Head of Football Data Analytics at RB Leipzig. - Grade/Score : (German) 1.2 ≈ (U.S. GPA) 3.8 / 4.0
BSc. (Hons) Physics, Astrophysics & Cosmology at Lancaster University
October 2, 2017 - June 26, 2020
▪ Quantitative knowledge in mathematical and physical modelling, statistics, and scientific computing – the building blocks for later work in ML, DL, and spatio-temporal football/sports models & visualizations: - Mathematics : Advanced Calculus · Advanced Linear Algebra · Advanced Statistics · Numerical & Complex Methods - Physics : Advanced Quantum Mechanics · Advanced Astrophysics & Cosmology · Special & General Relativity · Particle Physics · Nuclear & Atomic Physics · Solid State Physics · Thermodynamics · Electromagnetism
A-Levels, AS-Levels, IGCSEs at Bellver International College
September 1, 2002 - June 23, 2017

Qualifications

Agent Evaluation on Databricks (and MLflow 3.x) - Accreditation Badge (Databricks)
August 10, 2026 - August 16, 2026
Building RAG Agents (Knowledge Assistants) with Agent Bricks - Accreditation Badge (Databricks)
August 3, 2026 - August 9, 2026
Introduction to LangSmith - Certificate (LangChain)
July 20, 2026 - July 26, 2026
AI Context Engineering for Football - Certificate (Twelve Football)
February 24, 2026 - May 29, 2026
Get Started with AI Agents on Databricks - Accreditation Badge (Databricks)
February 2, 2026 - February 8, 2026
Introduction to LangGraph - Certificate (LangChain)
October 25, 2025 - October 31, 2025
AI Agent Fundamentals - Accreditation Badge (Databricks)
October 18, 2025 - October 24, 2025
Generative AI Engineering - Certificates (Databricks)
October 1, 2025 - October 12, 2025
Generative AI Fundamentals - Accreditation Badge (Databricks)
September 27, 2025 - September 30, 2025
Machine Learning Engineering - Certificates (Databricks)
July 19, 2025 - July 27, 2025
GCP Machine Learning: Vertex AI | Gemini | Generative AI - Certificate (Udemy)
July 1, 2025 - July 13, 2025
Building Agentic Applications on Databricks - Accreditation Badge (Databricks)
September 5, 2026 - September 5, 2026
Deploying & Monitoring Agent Applications on Databricks - Accreditation Badge (Databricks)
September 6, 2026 - September 6, 2026

Industry Experience

Other, Professional Services, Software & Internet, Media & Entertainment, Healthcare
    New End-to-End (Databricks-inspired) Football Data, Analytics & AI Platform

    ▪ Designing and engineering an end-to-end, Databricks-inspired, full-stack Football Data, Analytics & AI Platform combining provider-adapter ingestion, RBAC & audited sharing, DuckDB governed data/model catalogues across 40 sections with capability- & data-gated features, and more. Supports Streamlit/FastAPI interfaces, optional MCP and a Next.js/WebSocket collaboration client.

    ▪ Implemented capabilities: football modelling, ML (+ MLOps) & RL experimentation, capability-gated DL & computer-vision (CV) prototypes, event-derived pitch-value & graph-based analytics, RAGs/GraphRAGs, AI agents, EvalOps, observability & monitorization, and more…

    ▪ In progress: tactical simulation, multimodal fusion, interactive Ops Canvas, a governed ball-impersonated copilot assistant christened “Tiki” (from the ‘tiki-taka’ term), and more on their way…

    #Football #Soccer #FootballAI #SoccerAI #SportsAI #FootballAnalytics #SoccerAnalytics #SportsAnalytics #AgenticAI #GenAI #AIContextEngineering #ContextEngineering #LLM #LLMs #AIPlatform #AIPlatforms #AIPlatformArchitect #AIPlatformAchitecture #AIArchitecture #AIArchitectures #AIArchitect #AIAchitecture #MultiAgenticSystems # AgenticSystems #DeepLearning #MachineLearning #DL #ML #AI #DLEngineering #MLEngineering #AIEngineering #GenAIEngineering #DataScience #NeuralNetworks #Transformers #ExpectedGoals #xG #Streamlit #App #ChatBot #ConversationalApp #ChatApp

    CBs & CB-Pairings AI Analyst (Agent)
    I got selected onto Twelve Football’s 1st edition of the course (AI) “Context Engineering for Football” (a very limited cohort of 45 people worldwide was hand-picked by Founder David Sumpter himself + his comittee) - implemented a hands-on project as part of the course, to develop an AI Analyst agent; I implemented a conversational CBs & CB-Pairings AI Analyst agent: grounding an LLM in structured football event data through context engineering, RAG, and tool-use techniques. It is then delivered as a Streamlit chat front-end app, answering natural-language football scouting and analyst questions with professional-analyst-grade player comparisons & pairing recommendations. AI Analyst Agents for other positions on the pitch are on their way... Football Soccer FootballAI SoccerAI SportsAI FootballAnalytics SoccerAnalytics SportsAnalytics AgenticAI GenAI AIContextEngineering ContextEngineering LLM LLMs AIPlatform AIPlatforms AIPlatformArchitect AIPlatformAchitecture AIArchitecture AIArchitectures AIArchitect AIAchitecture MultiAgenticSystems AgenticSystems DeepLearning MachineLearning DL ML AI DLEngineering MLEngineering AIEngineering GenAIEngineering DataScience NeuralNetworks Transformers ExpectedGoals xG Streamlit App ChatBot ConversationalApp ChatApp…
    New Sports AI & Analytics Business Line

    As a side-project, independent from my daily duties, I single-handedly designed and built successfully from scratch a new business line for my department on Sports AI & Analytics, with special focus on football/soccer, under the sponsorship and direct supervision of a department-linked share-holding Partner.

    He supported carrying out this initiative after I proactively proposed it, shortly after joining EY, pitching the potential of this idea and business opportunity to the entire department.

    Shot Classification & Goal-Probability (xG) Estimation Using Graph Neural Networks (HGTs)
    ▪ State-of-the-art and cutting-edge MSc. Data Science thesis’ research project – modelling football shots as heterogeneous graphs of all players, the ball, as well as the instantaneous spatial & physical context, and then applying Heterogeneous Graph Transformers (HGTs) to classify shot outcomes and estimate their respective goal probability (i.e. xG = Expected Goals). ▪ Successfully demonstrated statistically that HGTs significantly improve the shot outcome classification, and their respective goal-probability estimation & calibration over traditional xG models. This thesis was directly supervised under Dr. Gabriel Anzer (Head of Football Data Analytics at RB Leipzig). ▪ In addition, this project was selected by Founder David Sumpter as 1 of 4 finalists for Twelve Football’s “Pitch to the Pros 3”. Football Soccer FootballAI SoccerAI SportsAI FootballAnalytics SoccerAnalytics SportsAnalytics DeepLearning MachineLearning DL ML AI DLEngineering MLEngineering AIEngineering DataScience NeuralNetworks GraphNeuralNetworks GNN GNNs HGT HGTs Transformers GraphTransformers ExpectedGoals xG…
    Expected Possession Value (EPV)-Added surfaces & contours – static
    Implemented EPV-Added surface + contour overlays, in static (.png) image format, from football spatio-temporal event & tracking data to visualize which areas of the pitch should the ball be displaced to, at every instant in time by the team in possession, to maximize their xG (Expected Goals) within that same passage play. Football Soccer FootballAI SoccerAI SportsAI FootballAnalytics SoccerAnalytics SportsAnalytics ExpectedPossessionValueAdded expectedpossessionvalueadded EPVAdded epvadded ExpectedPossessionValue expectedpossessionvalue EPV epv ExpectedGoals xG static png image explainerimage DeepLearning MachineLearning DL ML AI DLEngineering MLEngineering AIEngineering DataScience…
    Expected Possession Value (EPV)-Added surfaces & contours – animated
    Implemented EPV-Added surface + contour overlays, in animated (.mp4) video format, from football spatio-temporal event & tracking data to visualize which areas of the pitch should the ball be displaced to, at every instant in time by the team in possession, to maximize their xG (Expected Goals) within that same passage play. Football Soccer FootballAI SoccerAI SportsAI FootballAnalytics SoccerAnalytics SportsAnalytics ExpectedPossessionValueAdded expectedpossessionvalueadded EPVAdded epvadded ExpectedPossessionValue expectedpossessionvalue EPV epv ExpectedGoals xG animatedvideo animated mp4 video explainervideo DeepLearning MachineLearning DL ML AI DLEngineering MLEngineering AIEngineering DataScience…
    Spearman Pitch Control surfaces – static
    Implemented Spearman’s Pitch Control surface overlays, in static (.png) image format, from football spatio-temporal event & tracking data to visualize the areas of the pitch that are under control by each of the teams and which areas would be under dispute for the ball. Football Soccer FootballAI SoccerAI SportsAI FootballAnalytics SoccerAnalytics SportsAnalytics Spearman PitchControl pitchcontrol ExpectedGoals xG static png visualization explainerimage DeepLearning MachineLearning DL ML AI DLEngineering MLEngineering AIEngineering DataScience…
    Spearman Pitch Control surfaces – animated
    Implemented Spearman’s Pitch Control surface overlays, in animated (.mp4) video format, from football spatio-temporal event & tracking data to visualize the areas of the pitch that are under control by each of the teams and which areas would be under dispute for the ball. Football Soccer FootballAI SoccerAI SportsAI FootballAnalytics SoccerAnalytics SportsAnalytics Spearman PitchControl pitchcontrol ExpectedGoals xG animated mp4 video explainervideo DeepLearning MachineLearning DL ML AI DLEngineering MLEngineering AIEngineering DataScience…