This startup wants to make enterprise software look more like a prompt

In the rapidly evolving landscape of technology, the integration of AI into the workplace is anticipated to reshape conventional software interfaces. Josh Sirota, the founder of the startup Eragon, envisions a future where traditional software elements like buttons and menus become obsolete. Backed by a recent $12 million funding round, Eragon aims to create an AI-driven operating system tailored for enterprise clients, boasting a post-money valuation of $100 million. Sirota's bold assertion that 'software is dead' encapsulates his mission: to allow businesses to operate through natural language prompts. Eragon seeks to consolidate key business applications—such as Salesforce, Snowflake, Tableau, and Jira—into a cohesive interface powered by large language models (LLMs). Transitioning from a career at Oracle and Salesforce to launching Eragon, Sirota has gathered a team in San Francisco, where they are developing a product designed to redefine how companies interact with software. In a recent demonstration at their workspace, equipped with a bottle of Moët and Mac minis, Sirota showcased Eragon’s capabilities. The startup employs open-source models trained on specific customer datasets, creating a seamless link to existing company resources. For instance, when onboarding a new client like Dedalus Labs, Sirota simply issues a natural language request. The system autonomously generates user credentials and initiates an onboarding process, highlighting its efficiency. Eragon's approach is not without challenges. Potential vulnerabilities in AI agents and the complexities of ensuring data security within corporate environments are significant concerns. Nevertheless, the company has already begun deploying its technology across select large enterprises and numerous startups. Nico Laqua, CEO of the insurance startup Corgi, praised Eragon as the leading applied AI solution for enterprises, emphasizing the importance of data security and ownership. Sirota believes that models trained on extensive corporate histories will become invaluable assets. Unlike centralized models accessed via API, Eragon's framework allows companies to retain control over their data and AI configurations. This local approach is likened to the shift from mainframes to personal computers, suggesting that widespread corporate adoption will hinge on tailored tools that meet specific business needs. The competitive landscape is heating up, with industry leaders like Nvidia's CEO Jensen Huang advocating for similar agentic AI solutions. Huang's insights at the recent GTC conference reflect the growing recognition of AI's transformative potential in business practices. As Sirota confidently anticipates Eragon's valuation to reach $1 billion by year's end, he acknowledges the challenges ahead but remains optimistic that Eragon will provide executives with the tools necessary to enhance operational efficiency.