DevRel newsletter — September 2026
Blog post from Elastic
Elastic’s September 2026 DevRel newsletter highlights new AI, search, observability, and security capabilities, including jina-ocr-v1, a layout-aware OCR model available through Elastic and other deployment options that converts diverse documents into structured Markdown across more than 100 languages. The technical-preview Elastic CLI provides consistent, schema-validated access to Elasticsearch, Kibana, and Elastic Cloud APIs, with keychain credential storage, command policies, confirmation safeguards, and streamlined outputs for agent use; Elastic Agent Skills now use it as their secure transport layer. The newsletter also presents Knowledge Indicators, compact LLM-generated document summaries stored in Elasticsearch AI Indices that reportedly reduce retrieval token use and latency compared with conventional full-document RAG, alongside APM enhancements that display GenAI prompts, responses, and token consumption within OpenTelemetry traces. Elastic Security Labs describes auditing AI coding agents through Cursor hooks, gathering 13 million events from more than 1,100 machines to analyze tool use, file access, and MCP server activity. Additional resources cover vector search, SOC automation, Lucene performance, dashboards, time-series data, and community projects, while upcoming Elastic{ON} events are scheduled in cities including Mumbai, New York, Amsterdam, San Francisco, London, and Singapore.
| Trend | Post Mentions | Total Month Mentions | Posts | Companies | MoM |
|---|---|---|---|---|---|
| LLM | 9 | 6,033 | 1,166 | 244 | +15% |
| AI Coding Assistant | 4 | 1,596 | 558 | 143 | +2% |
| MCP | 4 | 10,589 | 923 | 216 | +14% |
| OpenTelemetry | 4 | 708 | 130 | 59 | -11% |
| RAG | 4 | 864 | 207 | 84 | -29% |
| Serverless | 4 | 1,224 | 281 | 103 | +39% |
| AI Agents | 2 | 5,609 | 1,414 | 272 | -6% |
| Kubernetes | 2 | 4,363 | 405 | 103 | +23% |
Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.