Experiments with OpenSearch and AI
summary
Proof-of-concept uses ChatGPT (and other LLMs) to convert natural language into Elasticsearch/OpenSearch DSL and also tests OpenSearch’s built‑in vector capabilities via the OpenSearch neural search plugin. Results show only a 33% hit rate for LLM-generated queries against provided index mappings and sample data, a failed small-scale fine‑tune due to insufficient training examples, and a working pipeline that indexes embeddings into a KNN index and demonstrates retrieval-augmented generation improving relevance by combining semantic retrieval with LLM responses.
AI-written summary · about
chapters (5)
- 0:00Introduction to experiments
- 2:30Natural language queries with LLMs
- 12:11Vector search and semantic search
- 36:35Future of AI in log analytics
- 37:19Q&A session
related talks
28mUse OpenSearch as Your Vector Search Database To Optimize GenAI Accuracy!
26mOpenSearch: Conversations with Context
30mBeyond Keywords: Building an AI-Powered Search That Reads Your Mind
51mQuery Understanding With Large Language Models: Techniques and...
22mAgentic Hybrid Retrieval: Orchestrating LLM Agents, Hybrid Search & Explainability in OpenSearch
42mOpenSearch With Gaia: Building Next-Gen AI-Driven Search and Analytics Workflows
47mBoosting AI With OpenSearch: Vector Stores, Similarity Search, and Retrieval-Augme...
49mLLM—Search Engine Killer?