Elastic Inference Service

elastic-vectorsearch-workshop is a public workshop repository containing instructor guidance, Elasticsearch queries, Python notebooks, corpus data, and agent setup code for practical vector, lexical, hybrid, and reranked search labs. It teaches semantic retrieval with Elasticsearch embeddings, BM25 search, and hybrid retrieval using reciprocal rank fusion and MinMax-normalized linear combinations, including judgment-set evaluation and weight sweeps. Later labs connect the retriever to an LLM and Elastic Agent Builder, demonstrating how retrieval quality affects generated answers, citations, and multi-hop agent workflows. The repository includes five lab notebooks, an Instruqt track, ingestion and validation scripts, and pre-event checks for corpus behavior, inference endpoints, and the workshop environment.

Elasticsearch is a distributed search and analytics engine, scalable data store, and vector database developed by Elastic. It stores and indexes structured or unstructured text, numerical, geospatial, log, and metric data as JSON documents, which clients submit and query through REST APIs or language clients. Documents are organized in indexes and data streams and become searchable in near real time; supported search modes include full-text, vector, and hybrid retrieval for applications such as logs, metrics, application performance monitoring, security analysis, and retrieval-augmented generation. It can be used as a managed deployment through Elastic Cloud or installed and operated independently. The repository also documents a Docker-based local setup with Elasticsearch and Kibana for development and testing, and Kibana provides tools for exploring indexed data, filtering it, and creating visualizations and dashboards.

Agents That Own Their Inference is a workshop resource for building and operating production AI agents on dedicated GPUs. Its Jupyter notebooks, Kubernetes manifests, load-testing utilities, evaluation code, and Terraform setup guide participants through running a vLLM OpenAI-compatible endpoint and measuring inference behavior, including prefill and decode, time to first token, token generation time, KV-cache capacity, batching, queueing, and saturation. The workshop establishes a baseline, changes serving configurations, reruns realistic load tests, and evaluates output quality alongside throughput and latency. Exercises cover BF16-to-FP8 quantization, speculative decoding with a draft model, vLLM serving-policy changes, dense versus mixture-of-experts behavior, and deploying an agent on the measured inference endpoint. It can use an Akamai-hosted workshop namespace or a participant's Kubernetes cluster with an NVIDIA GPU; the repository is licensed under Apache-2.0.

A serverless AI inference platform from Cloudflare that runs machine-learning models on Cloudflare's network and exposes them through the Workers platform. It supports tasks such as text generation, image generation, speech recognition, and text embeddings, making it usable for applications including semantic search.

Elastic Agent Builder is an agentic platform for defining agents with skills, tools, and workflows. Its API-driven endpoints can be reused in custom applications, and its demonstrations connect agents to Elasticsearch retrieval, search tools, ES|QL, and answers with citations to underlying documents. It is part of Elastic's platform for search and AI, which provides vector and hybrid search, reranking, inference, and agentic orchestration across cloud, on-premises, and serverless deployments.