cozymori
cozymori

Simpler, Easier, For Developers. Open-source frameworks for AI observability.

Products

  • VectorWave
  • VectorSurfer

Resources

  • Documentation
  • GitHub

© 2026 cozymori. All rights reserved.

Built with simplicity.

VectorWave

Vectorizing
framework for AI

A decorator-based Python framework. Vectorize function calls, store them in Weaviate, and get semantic caching, self-healing, and drift detection.

Get StartedSource Code
Scroll
VectorWave

One decorator is all you need

Trace and cache your functions with a single decorator. Supports both sync and async.

📄MIT License🐍Python 3.10–3.13🔷Weaviate Powered🏷️v0.2.9
app.py
from vectorwave import vectorize, initialize_database

initialize_database()

@vectorize(
    semantic_cache=True,
    cache_threshold=0.95,
    capture_return_value=True,
    team="ml-team"
)
async def generate_response(query: str):
    return await llm.complete(query)

# First call: executes normally, stores in Weaviate
result = await generate_response("explain transformers")

# Similar query: returns cached result in ~0.02s
result = await generate_response("what are transformers?")

Features

Everything you need for AI function observability.

Semantic Caching

Similar inputs are detected via vector similarity and return cached results instantly. Drastically cut LLM costs and latency.

Self-Healing

When errors occur, AI analyzes logs and source code, generates a fix, and opens a GitHub PR automatically.

Rust Batch Engine

PyO3-based Rust native extension handles batch writes and sensitive data masking asynchronously.

Distributed Tracing

trace_id and span_id based distributed tracing. Track function call chains hierarchically.

Drift Detection

Monitor input vector distance from Golden Dataset centroid. Instant webhook alerts on anomalies.

RAG Search

Natural language Q&A over stored code and trace logs. Explore your codebase with AI.

📄MIT License
🐍Python 3.10–3.13
🔷Weaviate Powered
🏷️v0.2.9

Start the wave

pip install vectorwave and start AI function observability in minutes.

$pip install vectorwave
Get StartedGitHub