Operations | Monitoring | ITSM | DevOps | Cloud

GPUs are everywhere powering LLM inference, model training, video processing, and more. Kubernetes is often where these workloads run. But using GPUs in Kubernetes isn’t as simple as using CPUs. You need the right setup. You need efficient scheduling. And most importantly you need visibility. This post walks through how to run GPU workloads on Kubernetes, how to virtualize them efficiently, and how Coroot helps you monitor everything with zero instrumentation or config.

Hyperparameter tuning is a critical step in optimizing large language models (LLMs). Parameters such as learning rate, batch size, weight decay, and number of training epochs can significantly affect convergence behavior and final model performance. While several approaches like grid search or random search are widely used, executing them manually is inefficient; especially when each training run is compute-intensive.

Datadog has long supported the monitoring of instrumented Go applications through our Go tracer v1. As the Go ecosystem has continued to mature, we’ve been hard at work collecting feedback and improving upon the tracer’s capabilities and usability features. We are now thrilled to announce the release of our Go tracer v2.0.0. This major update includes better security and stability, and a new and simplified API.

Over the past decade, organizations have embraced automation in waves – starting with basic task scripts and Robotic Process Automation (RPA), then moving to hyperautomation, and now exploring “agentic AI” as the next frontier. Each step in this evolution has expanded the scope of what can be automated, and revealed new challenges. This blog offers a detailed comparison of RPA, hyperautomation, and agentic AI, their key differences, strategic advantages, and potential drawbacks.