Open-source projects for AI-powered developer tooling and soccer coaching analytics, maintained by @karstenskyt.
luxury-lakehouse — A serverless soccer analytics platform built on Databricks Lakebase — replacing a 6-service traditional AWS pipeline with a unified lakehouse architecture that scales to zero.
silly-kicks — Soccer analytics library — multi-provider SPADL + tracking conversion, VAEP/xT valuation, and spatial features (pitch control, cover shadows, player influence).
mad-scientist-skills — Claude Code plugin — 8 skills for architecture, security, observability, optimization, documentation, cognitive UX, and C4 diagramming.
pining-for-the-data — Open-source soccer tracking data — redistribution, validation, and mock provider API.
ruthless-efficiency — A general optimisation/search substrate: a pure hexagonal core + pluggable search strategies + pluggable compute backends.
A FHIR-inspired schema and microservice pipeline for structured soccer session plans.
| Repository | Description |
|---|---|
| osti | Core schema — Pydantic V2 models, JSON Schema, versioned data standard |
| osti-segment | PDF segmentation microservice — spatial regions, text extraction, diagram cropping via VLM |
| osti-parse | Text parsing microservice — segmentation output to structured OSTI SessionPlan |
| osti-render | Visualization CLI — annotated page analysis PNGs from segmentation and parse output |
sports-coaching-skills — MCP platform for coaching session validation with semantic search, tactical diagram rendering, and evaluation against configurable coaching frameworks.