April 14, 2026
4 curated links on AI agents, Python, security, Google Cloud, and engineering, selected by Médéric Hurier.
-
Deploy Gemma 4 on Cloud Run: Pay Only When You Actually Use It
medium.com
This guide explains how to deploy the new Gemma 4 model family on Cloud Run with GPU support, enabling cost-effective "scale-to-zero" inference using vLLM and Run:ai Model Streamer.
-
Anthropic faces user backlash over reported performance issues in its Claude AI chatbot
fortune.com
Anthropic is facing criticism from developers reporting a decline in Claude's performance, allegedly due to "effort" adjustments made to save on compute costs amidst an industry-wide shortage.
-
Build a robust and cost-effective gen AI strategy
cloud.google.com
Google Cloud recommends a hybrid strategy combining Provisioned Throughput for critical workloads and Pay-as-You-Go for variable traffic to optimize both cost and performance.
-
Towards developing future-ready skills with generative AI
research.google
Google Research introduced Vantage, a new AI experiment that uses simulated conversations with AI avatars to assess and develop high-level human skills like critical thinking and collaboration.