Corporate RAG AI Assistant
An intelligent conversational assistant powered by Retrieval-Augmented Generation that transforms how visitors interact with the Elinext website. The system automatically indexes site content, performs hybrid BM25 and semantic search across a knowledge base, and generates precise answers with source citations — all streamed in real time through a modern chat interface.
OpenAI
React
Python
FastAPI
Docker
About the project
The Product: An intelligent AI-powered navigation and search solution designed to help visitors explore Elinext's corporate content through a convenient conversational interface.
What it Does: The assistant helps users find relevant information about Elinext services, technologies, expertise, and portfolio. Rather than browsing multiple pages, visitors can ask questions and receive AI-generated responses supported, where available, by relevant source links.
How it Works: Built on a modern Python and React architecture, the solution uses hybrid search that combines semantic understanding, keyword matching, and AI-driven reranking to retrieve relevant information from the Elinext knowledge base. The assistant uses this retrieved context to help generate useful and traceable responses in real time.
Important Notice: AI-generated responses are provided for general informational purposes only. While the system is designed to improve relevance and provide source references where available, responses may be incomplete, inaccurate, or not up to date. Users should verify important information with an authorised Elinext representative.
Commercial and Contractual Information: Chatbot responses do not constitute a quotation, offer, discount, contract, project commitment, delivery commitment, or professional advice. Prices, availability, project scope, timelines, and contractual terms are valid only when confirmed in writing by an authorised Elinext representative.
The Advantage: The solution provides a scalable conversational interface for navigating corporate information, with real-time response streaming, source-aware retrieval, and quality-evaluation processes. It is designed to help visitors access information more efficiently while maintaining clear paths to human consultation for important business decisions.
About the project
The Product: An intelligent AI-powered navigation and search solution designed to help visitors explore Elinext's corporate content through a convenient conversational interface.
What it Does: The assistant helps users find relevant information about Elinext services, technologies, expertise, and portfolio. Rather than browsing multiple pages, visitors can ask questions and receive AI-generated responses supported, where available, by relevant source links.
How it Works: Built on a modern Python and React architecture, the solution uses hybrid search that combines semantic understanding, keyword matching, and AI-driven reranking to retrieve relevant information from the Elinext knowledge base. The assistant uses this retrieved context to help generate useful and traceable responses in real time.
Important Notice: AI-generated responses are provided for general informational purposes only. While the system is designed to improve relevance and provide source references where available, responses may be incomplete, inaccurate, or not up to date. Users should verify important information with an authorised Elinext representative.
Commercial and Contractual Information: Chatbot responses do not constitute a quotation, offer, discount, contract, project commitment, delivery commitment, or professional advice. Prices, availability, project scope, timelines, and contractual terms are valid only when confirmed in writing by an authorised Elinext representative.
The Advantage: The solution provides a scalable conversational interface for navigating corporate information, with real-time response streaming, source-aware retrieval, and quality-evaluation processes. It is designed to help visitors access information more efficiently while maintaining clear paths to human consultation for important business decisions.
Features
Hybrid Search Engine
At the core of the chatbot lies a dual search engine that combines two complementary retrieval methods. BM25 full-text search via Tantivy delivers fast, precise document retrieval based on exact keyword matches with phrase boosting — ideal for specific queries with technical terms. Dense semantic search via Faiss uses OpenAI embeddings (text-embedding-3-large) to find relevant content even when there are no direct text overlaps between the query and the source material.
Results from both methods are merged through hybrid ranking with configurable weights and then refined by a reranking stage — either a local cross-encoder model (BAAI/bge-reranker-base) or LLM-based reranking via API. This multi-layered approach ensures that the final result set is both comprehensive and precisely ordered by relevance, delivering the highest quality context for answer generation.
Intelligent Query Pipeline
Every user question passes through a sophisticated multi-stage pipeline before an answer is generated. First, an LLM-powered routing layer determines whether the query requires a knowledge base lookup or can be handled directly — saving resources on small talk and off-topic requests. Next, the original question is rewritten with conversational context to produce an optimized search query that captures the user's true intent.
After the hybrid search returns the most relevant content chunks, the LLM synthesizes a natural language answer grounded in the retrieved sources. Each response includes inline source citations and a confidence score, giving users full transparency into where the information comes from. The entire conversation history is persisted for multi-turn dialogue support, enabling follow-up questions that build on previous context.
Production-Grade Architecture
The system is engineered for real-world reliability from the ground up. The LLM client supports primary and fallback API endpoints with automatic switching, ensuring uninterrupted service even during provider outages. When structured output parsing fails, a plain text fallback keeps the conversation flowing. All external dependencies — including Langfuse for observability, and third-party APIs — operate in graceful degradation mode, meaning their unavailability never blocks core chat functionality.
Responses are streamed to the frontend in real time via Server-Sent Events (SSE), providing an instant, fluid chat experience. The entire application is containerized with Docker using multi-stage builds, orchestrated via Docker Compose, and deployed through GitHub Actions CI/CD with automated linting, tests, builds, and health checks. A dedicated quality evaluation framework with benchmark datasets and A/B configuration comparison enables systematic, data-driven improvement of answer quality over time.
Features
Hybrid Search Engine
At the core of the chatbot lies a dual search engine that combines two complementary retrieval methods. BM25 full-text search via Tantivy delivers fast, precise document retrieval based on exact keyword matches with phrase boosting — ideal for specific queries with technical terms. Dense semantic search via Faiss uses OpenAI embeddings (text-embedding-3-large) to find relevant content even when there are no direct text overlaps between the query and the source material.
Results from both methods are merged through hybrid ranking with configurable weights and then refined by a reranking stage — either a local cross-encoder model (BAAI/bge-reranker-base) or LLM-based reranking via API. This multi-layered approach ensures that the final result set is both comprehensive and precisely ordered by relevance, delivering the highest quality context for answer generation.
Intelligent Query Pipeline
Every user question passes through a sophisticated multi-stage pipeline before an answer is generated. First, an LLM-powered routing layer determines whether the query requires a knowledge base lookup or can be handled directly — saving resources on small talk and off-topic requests. Next, the original question is rewritten with conversational context to produce an optimized search query that captures the user's true intent.
After the hybrid search returns the most relevant content chunks, the LLM synthesizes a natural language answer grounded in the retrieved sources. Each response includes inline source citations and a confidence score, giving users full transparency into where the information comes from. The entire conversation history is persisted for multi-turn dialogue support, enabling follow-up questions that build on previous context.
Production-Grade Architecture
The system is engineered for real-world reliability from the ground up. The LLM client supports primary and fallback API endpoints with automatic switching, ensuring uninterrupted service even during provider outages. When structured output parsing fails, a plain text fallback keeps the conversation flowing. All external dependencies — including Langfuse for observability, and third-party APIs — operate in graceful degradation mode, meaning their unavailability never blocks core chat functionality.
Responses are streamed to the frontend in real time via Server-Sent Events (SSE), providing an instant, fluid chat experience. The entire application is containerized with Docker using multi-stage builds, orchestrated via Docker Compose, and deployed through GitHub Actions CI/CD with automated linting, tests, builds, and health checks. A dedicated quality evaluation framework with benchmark datasets and A/B configuration comparison enables systematic, data-driven improvement of answer quality over time.
Contact Us
Anastasia Timoshenko
Regional Account Manager
1000+
Delivered projects
300+
Clients worldwide
700+
In-house developers
28+
Years in industry
5.0 Ratings