Pathway is a reactive data processing framework which lets you build
elegant and ultra-powerful real-time data products.
Code in Pathway feels familiar to data scientists and engineers alike. You can program in Python or SQL.
When you are done with batch mode, you connect live data sources, and simply run in streaming mode.
Pathway lets you easily design Machine Learning features, enrich your data, and draw insights quickly. You can use it for graph models, time series, geospatial data, and more.
Pathway is powered by a distributed pure-Rust streaming engine. It achieves high throughput with astonishingly low latency.
Pathway also preserves ACID-level data consistency.
Unlike streaming frameworks which only promise eventual consistency, Pathway guarantees data consistency at all times.
Pathway revises data outcomes, as needed. You can delete or update past data inputs that need changing, and include late data points.
Pathway will update your results and models reactively, with the minimal computation necessary.
Reactivity means less hassle with model versioning and less retraining following user feedback or data quality issues.
Pathway is currently used for real-time anomaly detection, predictive analytics, IoT and logs data observability, recommender systems, and alerting.
You can also use Pathway simply to reduce cloud costs, when frequent batch recomputation is eating away at your data budgets.
The data processing layer of Pathway is free and source available.
Market Guide for Event Stream Processing
Market Guide for Data Analytics and Intelligence Platforms in Supply Chain
with smart tables computed live by Pathway
using Pathway on top of a Kafka real-time data pipeline
to unlock Pathway-powered analytics for your ELK stack for free
and transform static and live data sources with Pathway