XERJ

XERJ is an AI-native search engine that indexes code, documents, and data for agents using a single binary.

XERJ is a unified search engine and vector database built specifically for AI agents, replacing multiple tools with a single Rust binary. It indexes code, documents, logs, and various file formats to answer agent queries directly from the index, which significantly reduces token consumption compared to traditional grep-driven workflows. The engine speaks the Elasticsearch API, allowing existing clients to connect without modification while using lexical hash-based embeddings for hybrid lexical and vector retrieval. It is designed for developers and enterprises building AI systems that require fast, scalable data access, isolated agent memory, and precise code definition retrieval. Supported inputs include thirteen format families such as JSONL, CSV, SQLite, PDF, DOCX, and YAML, running via a REST API.

XERJ at a glance

Best for: Software engineer
Runs on: Linux, macOS, Windows, self-hosted, open source, API
Pricing: Community edition is free; Professional is priced per-node; Enterprise is custom pricing.
Released: June 2026
Category: AI search engine
Website: xerj.org

Key features

One-command autoindexing for folders and files
Elasticsearch API compatibility
Built-in lexical and vector retrieval
Isolated agent memory namespaces
Precise code definition search
Resumable and idempotent data ingestion
Static binary deployment

Use it for

Indexing codebases for coding agents
Querying documents without context window waste
Storing and recalling agent memories by meaning
Running log analytics and anomaly detection

Alternatives to XERJ

Exa — Provides high-accuracy web search and deep research endpoints for AI agents.
OpenResearch — Run parallel research workflows locally or on remote compute.
CocoIndex — Maintains fresh context for AI agents with minimal incremental processing.
memsearch — It shares continuous context across Claude Code, Codex, and other AI coding agents.
Orama — Runs in browsers, servers, and edge networks with zero dependencies.
Searchbase — Independent web access with support for REST, MCP, self-hosting, and managed cloud.
Firecrawl — Provides direct web data access for AI agents with open source flexibility.
Engram — Single Go binary with zero dependencies and built-in full-text search.
USearch — It uses SIMD instructions and optimized concurrency to outperform FAISS.
Chemical X — Stops wasting agent context on CLI noise and compiler dumps.
Entire CLI — Mirrors Git repositories globally to increase cloning speeds for AI agents.
seekdb — Unifies relational, vector, and text data for production-grade AI agents.

All XERJ alternatives