Required stack
필요 기술
PythonSQLQlikViewTableauPower BIPlotlyDockerAWSSparkREST APIGraphQLData WarehousingETLData ModelingMachine LearningFastAPI
Project brief
프로젝트 내용
I’m building a data-driven application that delivers complete game-level and player-level analytics for four sports: Football, Cricket, Formula One, and Tennis. The goal is to merge historical and live feeds into a unified warehouse, run advanced modelling on top of it, and expose the results through interactive dashboards and an API other products can tap into.
Scope of work
• Set up or connect to reliable data sources, normalise them, and design an extensible schema that copes with multiple sports and rule sets.
• Develop the ETL pipeline (Python, SQL, Spark or a comparable stack) to ingest, clean, and enrich the data in near-real-time.
• Build an analytics layer capable of producing player performance metrics, match summaries, and cross-sport comparisons. Predictive models for injuries, form slumps, or race outcomes are a plus.
• Create a lightweight front end or embed visualisations (Tableau, Power BI, Plotly, etc.) so users can explore the numbers intuitively.
• Package the solution with clear documentation and deployment scripts (Docker / AWS preferred) so I can spin up additional environments easily.
Acceptance criteria
1. A populated database containing at least one full season of data for each sport listed.
2. Reproducible pipeline that updates automatically once new matches, races, or tournaments finish.
3. Dashboard and REST/GraphQL endpoint returning key metrics with sub-second latency on a modest cloud instance.
4. Clean codebase with README, setup guide, and inline comments.
When you respond, focus on your experience delivering multi-sport or high-volume analytics systems—tech buzzwords alone won’t suffice. If you have live demos or repositories you can share after signing an NDA, let me know; for now a concise overview of your hands-on experience is all I need to shortlist you.
Scope of work
• Set up or connect to reliable data sources, normalise them, and design an extensible schema that copes with multiple sports and rule sets.
• Develop the ETL pipeline (Python, SQL, Spark or a comparable stack) to ingest, clean, and enrich the data in near-real-time.
• Build an analytics layer capable of producing player performance metrics, match summaries, and cross-sport comparisons. Predictive models for injuries, form slumps, or race outcomes are a plus.
• Create a lightweight front end or embed visualisations (Tableau, Power BI, Plotly, etc.) so users can explore the numbers intuitively.
• Package the solution with clear documentation and deployment scripts (Docker / AWS preferred) so I can spin up additional environments easily.
Acceptance criteria
1. A populated database containing at least one full season of data for each sport listed.
2. Reproducible pipeline that updates automatically once new matches, races, or tournaments finish.
3. Dashboard and REST/GraphQL endpoint returning key metrics with sub-second latency on a modest cloud instance.
4. Clean codebase with README, setup guide, and inline comments.
When you respond, focus on your experience delivering multi-sport or high-volume analytics systems—tech buzzwords alone won’t suffice. If you have live demos or repositories you can share after signing an NDA, let me know; for now a concise overview of your hands-on experience is all I need to shortlist you.