A full-stack application combining FastAPI, Pandas, SQLite, Markdown/HTML reports and a Vue 3 dashboard. The system scores datasets, evaluates columns and turns quality rules into a professional visualization.
- FastAPI
- Pandas
- SQLite
- Vue 3
- TypeScript
- Vitest
- Docker
Context and problem #
Data teams need a quick reliability read before modelling, integrating or exposing datasets. The project automates that triage without hiding the applied rules.
My contribution #
The case highlights API design, a Pandas engine, persistence, tests and a decision-oriented interface with exports for technical documentation.
Architecture #
The Vue frontend sends the dataset to FastAPI, which runs quality rules, calculates scores, saves history in SQLite and generates HTML/Markdown reports.
- ✓ Rule-based engine
- ✓ Dataset and column scores
- ✓ Markdown/HTML export
- ✓ Optional LLM summary
Evaluation and quality #
Evaluation is oriented around observable problems: missing values, divergent types, distribution, duplication and actionable recommendations.
Limitations #
The system analyzes CSV files and aggregate metrics; it does not replace business validation or formal data contracts.
Next steps #
Add Excel, relational connections, run comparisons, quality alerts and PDF export.
Test the product and inspect the code
The live demo runs the core flow with safe local data. The public repository preserves the complete backend, tests, infrastructure and technical documentation.