AI-driven Progress Software, known for its innovative technology, has recently introduced Progress Agentic RAG, a SaaS Retrieval-Augmented Generation platform tailored for UK financial services.

This launch is expected to democratize access to reliable and verifiable generative AI across various organizations and teams. It signifies a significant step towards affordable AI development that can harness the potential of Large Language Models (LLMs) while ensuring results are grounded in practical business data.

Moreover, Progress Agentic RAG aims to bolster Progress Software’s comprehensive suite of end-to-end data management solutions, which include advanced data retrieval and contextualization features. This will enable businesses to unlock valuable insights from their diverse datasets, thereby gaining a competitive advantage. The financial firm is expected to maintain its commitment to adapting to market changes while adhering to stringent regulatory standards.

Details about Progress Software’s Launch of Progress Agentic RAG

To address the challenges posed by the growing volume of unstructured and structured data in different languages and formats, organizations need efficient ways to derive actionable insights. Simply relying on generative AI without context from all available business data can yield inaccurate and unusable information. Additionally, traditional Retrieval-Augmented Generation (RAG) solutions often demand substantial expertise and resources for deployment and operation.

Progress Agentic RAG is designed to offer a user-friendly and cost-effective approach to GenAI-powered search with streamlined setup and usage. Key features of the platform include:

• No-Code RAG Pipeline: A simplified ingestion process that indexes and retrieves data from multiple languages, text, audio, video, and other formats.
• Intelligent Search: Delivers AI-driven search capabilities on unstructured data to provide reliable answers.
• Deploy AI Agents: Specifically designed for scalable retrieval functionalities in AI agents.
• Multi-Model Integration: Supports all enterprise-ready Large Language Models (LLMs), allowing users to choose their preferred model.
• Purpose-built Database for RAG: Backed by NucliaDB, which provides semantic search, keyword search, metadata search, knowledge graph traversal, and multimodal understanding to generate coherent and trustworthy answers.
• RAG Evaluation Metrics: REMi metric ensures consistent answer quality and traceability.