#opensearch #opensource #opensearchambassador #opensearch #agenticai | Dotan Horovits

Hour 1: "I'll just spin up a quick search prototype." Hour 6: Still wrestling with ingest pipelines, cluster configs, and ML model deployment. 🥴 Sound familiar? If AI agents can write our code, why can't they configure our search infrastructure too? That's exactly what our new 𝗢𝗽𝗲𝗻𝗦𝗲𝗮𝗿𝗰𝗵 𝗟𝗮𝘂𝗻𝗰𝗵𝗽𝗮𝗱 does! Tell the AI agent what you want to search for, in natural language. The agent then collects a sample document, gathers preferences, plans a search architecture, and executes the plan (creating indices, ML models, ingest pipelines, and a local search UI). From zero to a working local search application in MINUTES (yes, minutes!) Without needing to be an #OpenSearch or search expert. And it's #opensource for all to enjoy! It works with Claude Code, Cursor, Kiro, OpenCode, Codex and any other agent that supports the Agent Skills specification (SKILL .md file). Just run: 𝚗𝚙𝚡 𝚜𝚔𝚒𝚕𝚕𝚜 𝚊𝚍𝚍 𝚘𝚙𝚎𝚗𝚜𝚎𝚊𝚛𝚌𝚑-𝚙𝚛𝚘𝚓𝚎𝚌𝚝/𝚘𝚙𝚎𝚗𝚜𝚎𝚊𝚛𝚌𝚑-𝚕𝚊𝚞𝚗𝚌𝚑𝚙𝚊𝚍 In fact, 𝗈𝗉𝖾𝗇𝗌𝖾𝖺𝗋𝖼𝗁-𝗅𝖺𝗎𝗇𝖼𝗁𝗉𝖺𝖽 is just the first skill, with more coming soon in OpenSearch Project, so stay tuned. Or better yet, contribute skills yourselves 😉 Huge shoutout to the Amazon Web Services (AWS) team behind this contribution - a joint effort of product, engineering, UX design and ML, to drastically simplify search setup. This is DevEx done right 👏 Owais Kazi Sarat Vemulapalli Krishna Kondaka Kaituo Li Janelle Arita Fen Qin Zhichao Geng Chenyang Ji ARJUN KUMAR GIRI Xiaoyu (Sean) Zheng Bobby Mohammed I've dropped the blog post and GitHub repo in the first comment below - check it out yourselves. What's your biggest pain point when setting up search infrastructure? And if you try Launchpad, I'd love to hear your experience! 💬 #OpenSearchAmbassador #OpenSearch #AgenticAI AWS Open Source