👋 Hi, I’m Ashish
🇮🇳 From India
🧑💼 Principal Solutions Architect - Search Specialist, Elastic
👀 Developer, Search Engineer (latest interest - Generative AI, vector databases, RAG, agents …)
🧑🤝🧑 Community lover, Speaker, OSS
👋 Hi, I’m Ashish
🇮🇳 From India
🧑💼 Principal Solutions Architect - Search Specialist, Elastic
👀 Developer, Search Engineer (latest interest - Generative AI, vector databases, RAG, agents …)
🧑🤝🧑 Community lover, Speaker, OSS
Hybrid Search Done Right: Stop Calling Metadata Filters “Hybrid” Everyone’s talking about hybrid search right now. But here’s the uncomfortable truth: 👉 Just because you glued vector search onto your database and added metadata filters doesn’t mean you’ve built true hybrid search. That’s like duct-taping a spoiler on a hatchback and calling it a race car. 🚗💨 Hybrid search is more than just “keyword + vector + filter.” It’s about field-level design, reranking, scoring, and scale. ...
🎤 Talk Summary: No-Code RAG Chatbot with PHP, LLMs & Elasticsearch Speaker: Ashish Diwali (Senior Developer Advocate, Elastic) 🔑 Introduction Topic: Integrating Generative AI (LLMs) with PHP. Goal: Show how to build chat assistants, semantic search, and vector search without heavy ML expertise. Demo focus: Using Elasticsearch + PHP + LLM (LLaMA 3.1). 🧩 Core Concepts 1. Prompt Engineering LLMs generate responses based on prompts → predicting next words. Techniques: Zero-shot inference → direct classification or tagging. One-shot inference → provide one example in the prompt. Few-shot inference → multiple examples → useful for structured outputs (SQL, JSON, XML). Iteration + context = In-context learning (ICL). 2. LLM Limitations ❌ Hallucinations (wrong answers). ❌ Complex to build/train from scratch. ❌ No real-time / private data access. ❌ Privacy & security concerns (especially in banking, public sector). 3. RAG (Retrieval-Augmented Generation) Solution to limitations. Workflow: User query → hits database/vector DB (e.g., Elasticsearch). Retrieve top 5–10 relevant docs. Pass as context window → LLM generates accurate answer. Benefits: Grounded responses. Works with private data. Avoids retraining large models. 🔍 Semantic & Vector Search Semantic Search: Understands meaning, not just keywords. Example: “best city” ↔ “beautiful city.” Vector Search: Text, images, and audio converted into embeddings (arrays of floats). Enables image search, recommendation systems, music search (via humming). Similarity algorithms: cosine similarity, dot product, nearest neighbors. 🛠️ Tools & Demo Elephant Library (PHP) Open-source PHP library for GenAI apps. Supports: LLMs: OpenAI, Mistral, Anthropic, LLaMA. Vector DBs: Elasticsearch, Pinecone, Chroma, etc. Features: document chunking, embedding generation, semantic retrieval, Q&A (RAG). Demo Flow Ingestion: ...
🚀 No-Code Chatbot with Elasticsearch + AWS Bedrock (Talk Summary) Speaker: Ashish (Senior Developer Advocate, Elastic) Event: AWS Community Day Mumbai 2024 🔑 Why Search Still Matters with LLMs LLMs (like ChatGPT) are powerful but face: ❌ Hallucinations 💰 High cost per query 🔒 No access to private / real-time data ✅ Search grounds LLMs in reliable, domain-specific info. ⚡ Elasticsearch Capabilities Traditional keyword search + modern vector search. Real-world use cases: 📍 Geospatial queries (ride-sharing, food delivery) ❤️ Matchmaking 📊 Observability dashboards 📝 Centralized logging (Elastic Stack: Elasticsearch, Kibana, Beats, Logstash) 🤖 Retrieval-Augmented Generation (RAG) Workflow: ...
This is a quick gist to demonstrate how you can run Elasticsearch securely with TLS on a public IP or domain. Sometimes, we need to spin up Elasticsearch for a development environment. In that case, you can follow these quick steps. Note - This gist is not recommended for production multinode cluster. As I only tested on single node cluster. If you have a multi-node cluster please follow the official guide. ...
Introduction Serverless architectures take on-demand tasks to the next level with event-driven scheduling of workloads. Elastic Observability gives you the same insights into your serverless activities as the rest of your environment. Gather logs and metrics from your serverless invocations and tie them together with traces from your serverless functions. For example Identify AWS Lambda latency issues, cold starts, and other invocation issues. Logs are collected with the rest of your telemetry data, so you can look at all your data in context, in one place. ...