D

duckdb

Options trading analytics — backtest stats, walk-forward analysis, Monte Carlo, risk metrics. Web dashboard at tradeblocks.io | MCP server on npm | Docker: romeo345/tradeblocks-mcp

DuckDB bindings for Kit - in-process analytical database

Documentation: https://oslandia.gitlab.io/qgis/qduckdb/

〽️ The eLxr Metrics project is designed to provide insights into the website performance and package distribution metrics. It tracks key indicators such as website view count and download statistics for binary packages📦, presenting the data in visually informative charts📊📈.

https://arrowpqt.gitlab.io

Arrow & Parquet is a high-performance, fully client-side data transformation and analysis tool designed to bridge familiar data formats with modern, high-speed querying engines directly in the browser. It allows users to import Arrow, Arrows, CSV, and JSON files, instantly converting them into the highly optimized Parquet format using Rust WebAssembly. Native Parquet files can also be imported directly.

Once the data is securely loaded, an embedded DuckDB WASM engine enables users to perform complex SQL queries, validate data structures, and aggregate large datasets with zero server latency. After filtering and shaping the data, the final output can be exported as either CSV or Parquet files for seamless integration into downstream machine learning workflows or BI tools.

To handle robust data operations and large WebAssembly payloads without hitting strict hosting limits, the application is deployed via GitLab Pages and utilizes the browser's Origin Private File System (OPFS). This guarantees high-speed local storage and ensures that all data processing remains strictly local, providing absolute data privacy without ever transmitting sensitive information to an external server.

Built with Astro, Rust WebAssembly, and DuckDB WASM.

Architecture warehouse retail (Kimball) — star schema complet avec dbt + DuckDB (dev) et Snowflake (prod). SCD Type 2, gouvernance des données, masquage PII et gestion des rôles.

Production-style real-time crypto streaming pipeline (Binance → Kafka → Polars Parquet → DuckDB → dbt) built locally with Docker and Python.

Remin is a local-first semantic search engine for personal notes. It combines embeddings, ANN vector indexing (HNSW), and reranking models to retrieve information based on intent rather than keywords — fully running on-device for privacy.

Built with Rust, DuckDB (VSS), and modern embedding models.

Local first, AI-powered data analytic workspace.