NewsAI in production
What Nano Banana 2.1 changes in an image-generation service
Google's new image model rejects familiar generation settings. Follow a thumbnail job through its outgoing request, retries and delivery to evaluate the upgrade.
Guides, news and explanations about AI and reliable systems.
NewsAI in production
Data Agent Kit brings Bigtable queries into coding agents
Bigtable joins Google’s Data Agent Kit. An illustrative rollout investigation shows why retained cell versions need an application-level evidence check.
NewsAI in production
Google API Gateway adds API-key authentication for MCP discovery
API Gateway now supports API keys for MCP tools/list. Discovery and execution keep separate authentication rules, and the MCP feature remains in Preview.
Cloud Tasks brings per-task retry settings to general availability
Cloud Tasks now lets each job override queue retry settings. A job’s expiry and the outcome of earlier attempts still determine whether another delivery is useful.
NewsObservability
Apigee adds model and MCP tool performance dashboards
Apigee separates model latency and token use from MCP tool traffic. Tool filters narrow a slow-assistant investigation, while proxy policies determine which charts populate.
Google API Gateway adds streaming for LLM responses
Google API Gateway’s streaming preview delivers response chunks while a model is still generating. Creation-time settings and request deadlines shape the migration.
NewsObservability
Google ends maintenance for legacy Monitoring and Logging agents
Standard maintenance has ended for Google’s legacy agents. Migration to Ops Agent changes some metric fields, including disk labels used by existing queries.
Storage Intelligence adds usage findings and batch dry runs
Storage Intelligence flags unusual storage usage and previews batch changes. A recurring dataset read shows why the job’s purpose matters when choosing which objects to move.
NewsAI in production
Google PageBreak tests AI-found bugs before sending reports
Google’s internal PageBreak agent proposes web vulnerabilities; separate validators test them in running applications. Reproductions can reduce the investigation work handed to developers.
AlloyDB previews separate compute for agent queries
AlloyDB’s preview gives read-only agent queries separate compute that scales to zero. Shared storage and the first query after idle time shape its value for incident investigation.
NewsAI in production
Antigravity brings local models to its agent SDK
Antigravity’s SDK supports Gemma through LiteRT and compatible local model servers. A hybrid cloud planner offers another arrangement, with filenames and task descriptions sent off-device.
GuideAI in production
Prompt caching: measure the cost of useful work
What prompt-cache hits save once retries and failed work are included in task cost, with a local calculator and ledger.