AI agents, also known as generative AI models, have gained recognition as potential productivity boosters due to their ability to autonomously perform tasks such as data entry and document processing. Orby AI, founded by Bella Liu and William Lu, is a company that is focused on building a generative AI platform to automate various business workflows.
While many startups and established players offer tools for automating back-office processes, Liu and Lu believe that Orby’s technology stands out because it can learn and act on workflows in real time, as well as understand the patterns and relationships within unstructured data.
Orby’s platform observes and learns from how workers perform their tasks and creates automations for complex tasks that require reasoning and understanding. An AI agent installed on a worker’s computer watches, learns, and generates automations, constantly adapting and improving as it learns more.
The goal of Orby is to abstract low-level decisions made by workers, freeing them up to focus on more strategic and critical tasks. Liu, with her background in AI and automation at IBM and UiPath, and Lu, with his experience as a systems engineer at Nvidia and engineering lead at Google Cloud, bring valuable expertise to Orby.
The secret sauce of Orby is its cloud-based generative AI model that is fine-tuned to complete specific customer tasks, such as validating expense reports. The model combines symbolic AI, which leverages rules and mathematical theorems, with traditional AI architectures to achieve scalability and efficiency. While symbolic AI alone can be inflexible, recent research has shown its potential when paired with other AI approaches.
Orby’s model is designed to intelligently adapt to changes in workflows, even when an application’s user interface is updated. It does this by analyzing API interactions and the worker’s browser usage. To address privacy concerns, Liu assures that Orby does not store customer data but rather uses it to improve its model. The data is encrypted during transit and at rest, and humans are kept involved in the feedback loop to ensure transparency and control.
In a competitive landscape, with generative AI powerhouses like OpenAI and Anthropic developing agentic AI solutions, companies like Orby face challenges. Adept, a startup focused on AI agents, is reportedly in talks with Microsoft for an acquihire deal, even before shipping its product. Tech giants like Amazon and Google have also released AI agent tools with limited attention. Despite the increased focus on generative AI, UiPath saw a decline in sales in its recent fiscal quarter.
Liu believes that Orby can succeed by taking a systematic approach to market. The company already has a dozen customers generating revenue, and with the recent $30 million Series A funding, they plan to expand their team and invest in go-to-market strategies, customer support, product development, and technical capabilities. The enterprise market is hungry for generative AI solutions that demonstrate tangible improvements in business performance, and Orby aims to provide that by identifying the best applications of the technology before scaling it across different businesses.
In conclusion, Orby AI is a company that is leveraging generative AI models to automate business workflows. Their platform, which learns from workers’ actions, creates automations that can handle complex tasks requiring reasoning and understanding. With a focus on real-time learning and adaptation, as well as the ability to understand unstructured data, Orby aims to free up workers to focus on more strategic tasks. With a systematic approach to market and a growing customer base, Orby has the potential to make a significant impact in the generative AI space.
Source link
