Polygres
AI retrieval platform for PostgreSQL databases, built by Evokoa
The founders
DP
The company
Polygres is an AI retrieval platform built on Postgres that unifies graph traversal, vector similarity, full-text search, and hybrid retrieval behind a single API. It brings Postgres, pgGraph, and pgVector together so developers can query meaning, relationships, and exact text matches without stitching separate systems or fragile ETL pipelines.
AI agents increasingly need to ask what happened, what something resembles, and who is affected — usually across a tangle of specialised databases. Polygres bets that teams would rather keep that intelligence inside the Postgres they already run than bolt on yet another stack.
Polygres is built by Evokoa, a research and engineering company with a stated mission to make the world's data AI-native. The founding team says it brings experience across infrastructure, AI, robotics, developer tools, scaling businesses, and community building.
Led by CEO Dalton Prescott Ng, CTO Damien Lim, and COO Dale Everett Ng, the team spans the infrastructure and AI ground this problem sits on, and is backed by Y Combinator. Early traction is with teams building enterprise agents, augmented reality, and healthcare AI.
Who else is building in this space
Companies working on the same thing, at every stage.
Autonomous truck technology for long-distance freight transport
AI agents automating supply chain supplier communication.
AI-native procurement services for businesses that make, move and serve physical goods
AI agent for multi-platform e-commerce inventory management
AI-powered unified supplier communication system for procurement
AI-native operating system for manufacturing plants, from purchase to payment
Thousands more founded every month.
Dealroom tracks new companies as they appear — founders, funding and signals, over the API or in the platform.
