Field notes — week of August 31, 2026

Anthropic released CLI 1.30.0 on Sept 3, introducing the 'ant apply' command that lets teams declare agents, environments, skills, memory stores, and deployments as files in a repository and reconcile them with a reviewable plan, writing a claude-lock.json lockfile for repeat runs. Salesforce restructured its product line into three tiers—Core, Advanced, and Max—on Sept 3, bundling AI, Slack, analytics, security, and support, making Agentforce native to every tier. Crowdin's August recap (Sept 4) highlighted a TypeScript CLI rewrite, four new automation Skills, an AI Pipeline builder in Copilot, and an Adobe Experience Manager connector, while Decagon published two engineering posts on Sept 3 detailing GPU-efficient inference serving that cuts GPU-hours by roughly 80% and an audio-native system for detecting mid-call speaker changes.

Field notes — week of August 31, 2026 What changed in AI-support tooling this week. Every item was checked against its source before it entered this list. - Anthropic - the ant apply command https://platform.claude.com/docs/en/cli-sdks-libraries/cli/apply CLI 1.30.0, Sept 3 lets teams declare agents, environments, skills, memory stores, and deployments as files in a repository and reconcile them with a reviewable plan, writing a claude-lock.json lockfile so repeat runs update the same resources instead of creating new ones. - Salesforce - the Winter '27 editions https://www.salesforce.com/news/stories/salesforce-simplifies-editions-2026/ Sept 3 restructure the product line into three tiers - Core, Advanced, and Max - that bundle AI, Slack, analytics, security, and support together, making Agentforce native to every tier instead of a separate add-on. - Crowdin - its August recap https://crowdin.com/blog/whats-new-at-crowdin-august-2026 Sept 4 covers a CLI rewritten in TypeScript for faster startup, four new automation Skills including one for i18n setup, an AI Pipeline builder in Copilot, and a new Adobe Experience Manager connector. - Lokalise - a blog post on AI translation quality https://lokalise.com/blog/ai-translation-quality-evaluation/ Sept 1 argues human review alone can't scale, and lays out a three-layer framework - pre-production assessment, in-production scoring, and post-edit analytics - for measuring it instead. - Decagon - two engineering posts Sept 3 : one describes disaggregating prompt processing from token generation https://decagon.ai/blog/gpu-efficient-inference-serving-stack to cut GPU-hours by roughly 80% for the same traffic; the other describes an audio-native system for detecting mid-call speaker changes https://decagon.ai/blog/audio-native-semantic-speaker-change-detection that combines speaker-embedding models with a multimodal LLM.