Skill UI

This skill provides comprehensive monitoring capabilities for GPU instances running on Vast.ai. It collects key metrics including GPU utilization, instance uptime, temperature, and accumulated costs. Use this when setting up operational monitoring dashboards, configuring critical alerts (e.g., idle GPUs, high temperature, budget overruns), or automating resource usage tracking in a cloud computing environment.

A comprehensive guide for optimizing query latency in Qdrant. This resource explains advanced performance tuning techniques, focusing on critical resource management like RAM utilization, segment count adjustments, and HNSW parameter tuning. It details strategies for scaling, quantization, and hardware considerations (e.g., local NVMe) to ensure consistently low-latency search performance, and outlines common pitfalls to avoid.

This skill generates a comprehensive capacity plan document for any service. It assists technical teams in proactively managing infrastructure by modeling traffic growth, forecasting resource utilization (CPU, memory, connections), projecting costs, and defining clear scaling strategies. Use this when assessing current resource headroom, planning for seasonal peaks, or defining a long-term infrastructure action roadmap.