·
Vector Database Founders
Andrew Kane
Creator & Maintainer |PGVector (open source)
Prolific OSS developer (430+ repos). Created pgvector — the default vector search extension for PostgreSQL.
GitHub
Open Source Projects
9 repositories · 85 total stars
iceberg-ruby
Apache Iceberg for Ruby
Rust341
ducklake-ruby
Run your own data lake with a SQL database and file/object storage
Ruby26
neighbor-s3
Nearest neighbor search for Ruby and S3 Vectors
Ruby13
seaduck
Apache Iceberg for Ruby, powered by libduckdb
Ruby7
torchcodec-ruby
Media encoding and decoding for Torch.rb
C++2
bayestest-c
Bayesian A/B testing calculations for C
C1
seaduck-php
Apache Iceberg for PHP, powered by libduckdb
PHP1
ducklake-php
Run your own data lake with a SQL database and file/object storage
PHP1
Languages
Tech Stack
Ruby33%C++22%PHP22%Rust11%C11%
Biography
Deep Research
Andrew Kane is the creator and maintainer of pgvector, the open-source PostgreSQL extension for vector similarity search that has become essential infrastructure for modern AI applications. He launched pgvector in 2022 to enable AI/ML applications to perform vector similarity search directly within PostgreSQL, leveraging the database's extensibility instead of introducing new specialized infrastructure. Kane maintains an exceptionally prolific open-source portfolio with 430+ repositories on GitHub, including PGHero (a PostgreSQL performance dashboard with 8.8k stars) and Searchkick (intelligent search for Ruby with 6.4k stars), demonstrating deep expertise in database tooling and developer productivity. In 2024, pgvector 0.6.0 introduced 30x faster parallel index builds and binary quantization delivering 150x speedups, cementing its position as the standard vector search solution across major cloud platforms including AWS RDS, Azure, Supabase, and Neon. Based in San Francisco, Kane has established himself as a key contributor to the AI infrastructure ecosystem through practical, production-ready database extensions.
PostgreSQL vector similarity search extension developmentHNSW index optimization for high-dimensional vectorsParallel index construction for vector databasesScalar and binary quantization methods for vector compressionPostgreSQL extension architecture and C-level developmentVector search performance benchmarking and optimizationDatabase-integrated machine learning infrastructureCross-language database bindings (Ruby, PHP, C/C++)Open-source database tooling and ecosystem developmentProduction vector database scaling and maintenance challenges
Timeline
4 GitHub4 Research8 total
2025
2025-10-01GitHub
Created seaduck: Apache Iceberg for Ruby, powered by libduckdb2025-08-27GitHub
Created iceberg-ruby: Apache Iceberg for Ruby2023
Key Contributions
PGVector
Open-source PostgreSQL extension for vector similarity search that enables AI/ML applications with vector embeddings directly in PostgreSQL. It transforms PostgreSQL into a production-ready vector database supporting exact and approximate nearest neighbor search with multiple distance metrics and indexing algorithms.
Searchkick
Intelligent search engine for Ruby on Rails that integrates with Elasticsearch/OpenSearch, providing advanced search capabilities including stemming, misspelling handling, synonyms, and personalized results. It's designed to learn from user behavior and improve search relevance over time.
PGHero
Performance dashboard and monitoring tool for PostgreSQL that provides real-time insights into database performance, query optimization, and system health. It helps developers identify slow queries, index opportunities, and database bottlenecks.
Chartkick
JavaScript charting library that creates beautiful, interactive charts with minimal code, integrating seamlessly with Ruby on Rails applications. It provides a simple API for creating various chart types while handling data formatting and responsiveness automatically.
Strong Migrations & Database Tooling Suite
Collection of database safety and productivity tools for Ruby on Rails including Strong Migrations (prevents unsafe database operations), Lockbox (field-level encryption), Ahoy (analytics), Blazer (business intelligence), and other essential database utilities.
Database Infrastructure Patterns
Established battle-tested patterns for database tooling and Ruby gem development that have been adopted as industry standards. His work demonstrates exceptional productivity with 431 public repositories and consistent architectural patterns across projects.
Vector Database Founders/Andrew Kane
All Profiles