返回
AI AgentDECAGON

Delivering high-performance customer support

Decagon 使用 OpenAI 多个模型构建自动化客服 Agent 平台

OpenAI News20 分钟阅读英文
阅读原文
Delivering high-performance customer support
TL;DR: Decagon 使用 OpenAI 多个模型构建自动化客服 Agent 平台
以下为 OpenAI News 原文(英文

OpenAILaunched in 2023, Decagon⁠(opens in a new window) has quickly become a key player in automating customer support for companies like Curology, BILT, Duolingo, Eventbrite, Notion, and Substack. OpenAI’s models are crucial in their ability to deliver fast, reliable responses—without human intervention.From enterprises to tech-forward startups, Decagon helps businesses globally handle millions of support conversations without sacrificing quality or speed. The company uses a combination of OpenAI’s models—including GPT‑3.5, 4, 4o, 4 Turbo, and OpenAI o1‑mini—to deliver agentic bots that go beyond response generation and service the entire customer lifecycle.Decagon’s customers require scalable, high-quality support that can handle complex inquiries. Their two founders, having successfully exited AI companies previously, recognized the need for a support solution that went beyond basic automation to deliver nuanced yet fast responses across vast numbers of interactions.“We know that latency has a direct impact on customer satisfaction. Every second counts when you're dealing with real-time customer support,” says Ashwin Sreenivas, Decagon’s co-founder and CTO.But maintaining a high level of automation while ensuring precision required more than just traditional automation tools—it required advanced AI models capable of reasoning through complex tasks. With OpenAI’s models, the Decagon team was able to architect a flexible solution that allows clients to fully customize their workflows to their needs.Decagon uses a creative combination of OpenAI’s GPT models, which offers flexibility and customization to meet the diverse needs of Decagon’s customers. OpenAI’s platform also helps the Decagon team optimize performance across a range of tasks: “We found that different models have different strengths,” Sreenivas notes.For example, Decagon fine-tuned GPT‑3.5 to rewrite customer queries before they enter retrieval-augmented generation (RAG) workflows. “We tried a bunch of other model configurations, and for customer queries, fine-tuning GPT‑3.5 got us the highest performance,” says Sreenivas. In other workflows, the company uses GPT‑4 for complex decision-making tasks, allowing the platform to process API requests and other intricate operations efficiently.Decagon’s approach ensures that each part of the customer interaction pipeline is supported by the most suitable model for the task, enhancing both speed and accuracy. Says Jesse Zhang, Decagon’s co-founder and CEO, “This allows us to both capture customers' business logic and create all the software surface area around the agent that just wasn't possible before LLMs.”By combining OpenAI’s models with Decagon’s tailored workflows, the company has dramatically improved the quality of customer service automation for its clients. “For one of our largest customers, we handle 91% of all their global support, without a human being involved,” says Sreenivas. The configurability of Decagon’s platform with OpenAI’s models allows it to adapt to each client’s specific requirements, ensuring highest performance and seamless integration. One key advantage is Decagon’s ability to rapidly evaluate and integrate new models. “Every time a model comes out, we can run them through evaluations really, really quickly,” says Sreenivas. This flexibility allows Decagon to stay on the cutting edge, ensuring that their platform is always running the most accurate and efficient models available.As Decagon continues to grow, it’s focused on expanding its capabilities to support an even wider range of customer needs. The company is already working with new versions of OpenAI’s models—critical in Decagon’s ability to deploy AI-powered customer service quickly. “For new customers, our core infrastructure can be up and running in days,” says Sreenivas.Looking ahead, Decagon is exploring how to further enhance its AI agents with voice capabilities, aiming to tackle voice-based customer support interactions with the same level of automation and accuracy.By continuously powering its platform with the latest OpenAI technology, Decagon is well-positioned to lead the next wave of innovation in AI-powered customer service.Keep reading