Ruder Finn Shifts to Custom LLM Model for its New Website

As a communications agency with a tech-first mindset, this move makes Ruder Finn among the first in the industry to make an LLM its primary interface

NEW YORK, Aug. 13, 2026 /PRNewswire/ -- Ruder Finn, a global communications agency known for pioneering what's next, today announced the release of its newly reimagined ruderfinn.com, featuring a large language model (LLM) as its primary user interface. The platform, designed as a living, tech-forward system, positions Ruder Finn as one of the first communications agencies to use an LLM at the center of its digital experience.

The new ruderfinn.com reshapes how audiences interact with agency content and expertise online: Rather than navigating static pages and a search bar, visitors can now engage in conversation, querying the agency's capabilities, services, and solutions through natural language. The site is powered by a custom AI system built on Anthropic's Claude models via Amazon Bedrock. It uses Retrieval-Augmented Generation (RAG) to ground responses in information retrieved through semantic and hybrid search, alongside a meticulously curated company knowledge graph—improving relevance, accuracy, and privacy.

The experience reflects the broader shift underway across digital ecosystems where conversational AI and LLMs are becoming the dominant source for how people access and understand information. The site does still include links to traditional pages for those that prefer accessing conventional web pages.

"We're watching interfaces evolve in real time," said Kathy Bloomgarden, CEO of Ruder Finn. "People don't browse anymore, they ask. If that's how the world is communicating, then our website shouldn't describe the future of communications. It should operate like it. That's the inspiration for the new ruderfinn.com, because at Ruder Finn, staying ahead of the market isn't just part of our philosophy, it's our practice."

"Our platform is designed to evolve with the AI ecosystem, seamlessly incorporating the most advanced models available," said Tejas Totade, CTO and Head of AI of Ruder Finn. "It's the same philosophy we bring to client work. The differentiator isn't the model, it's the system design, the data strategy, and the human expertise layered on top. We aren't the only agency doing this kind of work for clients, but we are just about the only one walking the technology, not just talking the technology, because we are a communications agency with a tech-company mindset."

The system architecture includes:

• A growing knowledge graph that maps Ruder Finn's capabilities, expertise, and engagement history. This layer is a highly structured internal database that organizes Ruder Finn's services, capabilities, people, and work into defined relationships, manually built and maintained by agency experts and strategists to provide semantic grounding while ensuring responses are accurate, contextual, and relevant. Client mentions are anonymized in line with client agreements.
• A "closed-loop" LLM that operates within a controlled environment. Inputs, outputs, and updates are optimized with AI but governed through human-in-the-loop workflows. The system uses Anthropic's Claude models as its core engine, with custom layers tailored to Ruder Finn's specific context and needs.
• A moderation and review layer that enables expert oversight, protects client confidentiality, and ensures quality control as the system evolves. The system uses a Retrieval-Augmented Generation (RAG) architecture to retrieve approved internal content at runtime, helping ensure responses remain grounded in authorized information and aligned with privacy and compliance requirements. Ruder Finn has intentionally prioritized human approval as a permanent layer of the system, reflecting the realities of enterprise AI and the sensitivity of communications data.
• An integrated CMS-to-vector pipeline roadmap, where approved content will begin to feed both traditional web pages and the vector index powering the retrieval layer.

The launch of the new ruderfinn.com mirrors how Ruder Finn itself operates as an organization, where technology merges with human expertise to amplify experiences and build stronger, more authentic connections.

About Ruder Finn

Ruder Finn is one of the world's largest independent global communications and integrated marketing agencies. Founded in 1948, Ruder Finn has defined and redefined PR for over 75 years, shaping communications that moves industry-defining brands, companies, and leaders from what's now to what's next. Ruder Finn provides clients with bold strategies based on a global perspective and localized market knowledge that redefine leadership, reimagine the marketplace, and rethink customer experiences. The agency is organized around five core areas of expertise—Healthcare, Technology, Commerce, Leadership and Workplace—with innovative tech incubator RF TechLab providing cutting-edge predictive analytics, AI-powered creative hub RF Studio53 offering breakthrough creative, and specialized teams delivering customer-focused digital engagement. Ruder Finn has offices across four continents including the U.S., Asia, Europe and the Middle East. Wholly owned agencies and offerings within Ruder Finn Group include: Ruder Finn Inc., RF Studio53, RF Engage, Touchdown, Peppercomm, RF Comunicad, Mantis, Flightpath, Atteline, Pandan Social, Missouri Creative, and The Equity Group. For more information visit www.ruderfinn.com.

Key Facts

• Company: Ruder Finn
• Announcement: Ruder Finn Unveils AI-Powered Website with a Custom LLM
• What's New: The website features a large language model (LLM) as its primary user interface, enabling users to engage with the agency through natural language conversation rather than just traditional navigation.
• Industry First: Ruder Finn is the first communications agency to deploy an LLM as the principal interface for content discovery and engagement on its website.
• Technology Approach:
  • Built on a closed-loop LLM
  • Operates in a closed-loop, moderated environment
  • Incorporates human-in-the-loop governance to protect client confidentiality and ensure accuracy
• Grounded Knowledge Retrieval:
  • Uses a carefully curated knowledge graph and semantic search to retrieve approved information at runtime, supporting accurate and relevant responses.
  • Goes beyond a case-study database or keyword search engine by understanding context and relationships.
• System Design:
  • Supported by a structured knowledge graph
  • Integrated with the agency's CMS, with a roadmap to feed approved content into vector-based retrieval
  • Includes moderation and review workflows
• Purpose: Aligns the website with how audiences now find information: by asking questions through conversational interfaces rather than browsing pages
• Positioning: Puts into operation the AI capabilities Ruder Finn delivers for clients, using its own platform as proof of concept.

View original content to download multimedia:https://www.prnewswire.com/news-releases/ruder-finn-shifts-to-custom-llm-model-for-its-new-website-302850906.html

SOURCE Ruder Finn