Simpler, Easier, For Developers.
Vectorize and trace every function call. Cache similar requests instantly. When errors occur, AI diagnoses them and opens a fix PR — automatically.
One decorator to vectorize and trace your functions. Semantic caching, self-healing, drift detection — an AI observability framework powered by Weaviate.
Learn morefrom vectorwave import vectorize
@vectorize(
semantic_cache=True,
cache_threshold=0.95,
capture_return_value=True
)
async def generate(query: str):
return await llm.complete(query)
# Similar queries → cached in ~0.02s
# Every call vectorized & traced to WeaviateExecutions
—Success
—Cache Hit
—Execution timelines, distributed traces, and AI-powered error diagnosis. Monitor all your AI function data in a real-time dashboard.
Learn moreNo complex setup — one decorator is all you need. Every API is designed to work right out of the box.
All our work is open-source. The best tools emerge when ideas flow freely and communities build together.
We obsess over developer experience. Focus on what you're building — we'll handle the rest.
From decorator to dashboard to testing. Everything connects.
Start tracing with @vectorize
Real-time observability dashboard
AI diagnoses and fixes errors
CLI regression testing for AI functions. Compare outputs with vector similarity and LLM judges.
Lightweight Streamlit dashboard. Start monitoring with just Python — no setup required.
All our projects are open-source. Read the docs or start contributing on GitHub.