Open Source

Read our blogs about open source projects like Apache Wayang or LSTEnergy, and related popular open source topics.

Introducing Scalytics Connect – Community Edition

Scalytics Connect – Community Edition is open-sourced under Apache 2.0, delivering the same high-throughput, low-latency AI inference platform that powers enterprise workloads. Clone the repo to run vLLM-driven chat and image APIs, vector services, and Live Search—all via an OpenAI-compatible endpoint. Self-host your AI stack, manage API keys, and extend every component to fit your use case.

Author:

Alexander Alten-Lorenz

The Return of Multiagent Systems - Thanks to LLMs

Multiagent systems (MAS) are groups of intelligent agents working together to achieve a common goal. LLMs power these agents, enabling them to remember, plan, and use tools. Examples include virtual simulations and scientific discovery agents.

Author:

Dr. Tristan Behrens

LST-E: Our Industry-Specific Energy Forecasting Model

LST-E is Scalytics' advanced energy forecasting model built on LSTM architecture, designed to optimize energy usage, reduce costs, and lower emissions. Trained using federated learning on decentralized data, it ensures compliance while delivering highly accurate, actionable predictions. Available on GitHub and HuggingFace, LST-E empowers businesses to make smarter energy decisions with tailored, regulation-compliant AI solutions.

Author:

Scalytics

Apache Wayang: More than a Big Data Abstraction

Apache Wayang is a powerful open-source framework for data platform processing, offering a flexible, cost-effective solution for federated data.

Author:

Dr. Jorge Quiané

Why is federated learning so crucial?

Federated Learning (FL) is revolutionizing distributed computing frameworks by enabling decentralized data processing and enhancing privacy. Early adopters like Google, Apple, and NASA use FL to improve products, increase operational efficiency, and maintain data privacy. This post delves into FL's role in the real-world adoption of distributed frameworks, its commercial applications, and the future impact on AI and data-driven industries.

Author:

Alexander Alten-Lorenz