Kryonexus

High-performance hybrid retrieval and ranking engine.

Vision

Combines lexical and semantic ranking for low-latency decision-grade results.

What problem it solves

Single retrieval methods miss relevance in noisy datasets.

Core superpower

• ⚡ Hybrid rank fusion with streaming result interfaces

Key use cases

• ✅ Search backends
• ✅ Decision support
• ✅ Facet-ready retrieval
• ✅ Relevance tuning

API surface

POST /search, WS /search, GET /health

Technical stack

• Core stack: FastAPI + BM25/FAISS strategy layer
• Runtime: local-first, self-hosted friendly
• Infra: Docker Compose + Caddy + Redis/Chroma/Ollama compatibility

Current status (Feb 2026)

• ✅ Public scaffold available
• ✅ Bilingual README (EN default + ES)
• ✅ CI + release baseline configured
• 🚧 Feature hardening in progress

Quick start

git clone https://github.com/smouj/Kryonexus.git
cd Kryonexus
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
python -m src.kryonexus.cli --help

Documentation

• Implementation Guide
• Architecture
• Runbook
• Deployment Guide
• Release Process
• Changelog

Contributing

Contributions are welcome. Please read CONTRIBUTING.md.

License

MIT © 2026 smouj

---

Other skills

Explore the full ecosystem here: smouj/smouj

Signature: @Smouj013