Football Data Engineer
Gemini Sports AI
Data engineering role building ingestion, standardisation and transformation pipelines for a football player trading platform. You would harmonise multi-vendor football datasets, resolve player and club entities, build performance KPIs, and put machine learning models into production with data scientists.
The application is with the employer, never with us. Posted 3 weeks ago. Added by hand 3 weeks ago.
The advert, from Gemini Sports AI
About Gemini Sports AI & The Role
Gemini Sports AI empowers football clubs worldwide with a competitive edge in player trading. Our source-of-record application turns disparate information streams into actionable player trading intelligence for sporting directors, recruitment heads, and data analytics teams.
We are seeking a Data Engineer (5 years experience) to own the ingestion, standardization, and transformation pipelines powering our platform. You will harmonize multi-vendor datasets, build performance KPIs, and deploy machine learning models to production alongside Data Scientists.
Apply here 🔽
Key Responsibilities
1. Multi-Club Data Ingestion & Harmonization (Batch Workflows)
- Build scalable, provider-agnostic ETL/ELT pipelines to ingest heterogeneous football data (event, tracking, reports, physical).
- Solve football entity resolution by mapping and reconciling player IDs, club identifiers, competitions, and contract metadata across external vendors.
- Implement automated schema validation, anomaly detection, and data contract checks for rigorous quality control.
2. Performance KPIs & Analytical Data Modeling
- Engineer clean, version-controlled transformations into medallion data layers (staging, enriched, analytics mart).
- Develop advanced football KPIs (e.g., possession-adjusted actions, pressure metrics, physical load, progression metrics) optimized for fast querying and historical comparisons.
3. Data Science & Model Production Support
- Operationalize predictive ML models (player valuation, similarity engines, recruitment fit scoring) into production pipelines.
- Build feature stores and reproducible pipelines ensuring training data strictly matches inference features.
4. Orchestration, Reliability & Platform Engineering
- Orchestrate workflows using tools like Airflow, Prefect, or Dagster aligned with matchweek and transfer window timelines.
- Maintain CI/CD practices (GitHub Actions), automated testing (dbt, Great Expectations), and infrastructure optimization.
Candidate Profile & Qualifications
Required Experience & Skills
- Track Record: 5 years of experience building reliable, production-grade batch data pipelines.
- Football Domain Knowledge: Strong understanding of football datasets across Event (Opta, StatsBomb) and Tracking/Physical data (SkillCorner, Opta Vision).
- Core Technical Skills: Advanced Python (pandas/polars/PySpark) and SQL optimization.
- Cloud Platforms: Modern cloud stacks (AWS, GCP, or Azure), Docker, and cloud data warehouses/lakehouses.
- Collaboration: Strong communication skills to effectively translate data science concepts into production-grade solutions.
Nice-to-Haves (Advantages, Not Mandatory)
- Databricks Ecosystem: Practical experience with Databricks (Delta Lake, PySpark, Auto Loader, Unity Catalog, or MLflow) is a strong plus.
- ML Deployment: Exposure to deploying ML models in batch pipelines or managing feature stores.
Working at Gemini Sports AI
- Direct, measurable impact on how professional clubs evaluate, value, and trade talent globally.
- A multidisciplinary environment bridging high-level software/data engineering with elite football analytics.
Shown here so it can be read in one place. The application is with the employer.
Missed this one? Get the next by email.
One email on the days a new role appears. Nothing on the days none does.
Not sure whether this one suits you, or want feedback on your application before it goes in? That is what the Recruitment Room is for.