Current topics, practical solutions, and fresh ideas: On the HMS blog, you'll learn how companies successfully use data and analytics.
Agent-to-Agent Protocol: When Companies Should Use Multi-Agent Systems
This article helps enterprise teams determine when A2A is the right level of collaboration, when MCP or APIs are sufficient, and what governance requirements must be met before implementation.
Why AI-Powered Code Migration Doesn't Scale Without Agent Systems
Individual LLMs deliver impressive results. However, when modernizing large legacy environments, analysis, orchestration, testing, and governance are what determine success. Read this article to find out why agent-based systems are key to this process.
Inside Forecasting: What Really Sets Strong Forecasting Applications Apart
Why Forecasting Is Rarely Just a Modeling Issue: This article explores the factors that determine the acceptance and value of forecasts in practice—from explainable AI to the integration of local data.
Explainable AI for Time Series Forecasts: Making Forecasts Explainable with SHAP
The article explains how Explainable AI makes time-series forecasts easier to understand. Using SHAP scores, forecasts are broken down into individual contributions, revealing the factors driving a forecast. The article also shows how forecast revisions between two forecast runs can be explained.
Erklärbare KI (XAI): Warum sie entscheidend für Vertrauen, Compliance und Business Impact ist
Erklärbare KI (XAI) macht KI-Entscheidungen nachvollziehbar und stärkt Vertrauen in automatisierte Systeme. Gleichzeitig unterstützt sie Compliance-Anforderungen wie den EU AI Act und sorgt dafür, dass KI in der Praxis wirklich genutzt wird.
How Do Companies Choose the Right Technology Partner? A Guide for CIOs
Choosing the right technology partner is a strategic decision for CIOs today. Learn which criteria really matter—from use-case fit to governance to implementation expertise.
Non-Generative AI Agents: When Classical AI Is Superior to Generative Models
The article explains when non-generative AI agents should be used and why traditional models remain superior to LLMs in many business-critical applications.
How can critical changes in the condition of intensive care patients be detected early and, at the same time, evaluated in a medically sound manner? A recent scientific paper published in *Communications Medicine*, a peer-reviewed journal in the Nature portfolio, addresses this question. Markward Britsch, a data scientist at HMS, was also involved in the work as part of an interdisciplinary research team.
