depthfirst, an applied AI lab focused on software security, has announced a $40 million Series A funding round as it positions itself at the center of the rapidly evolving cybersecurity landscape. The San Francisco–based company is developing what it calls General Security Intelligence, an AI-native platform designed to detect, triage, and remediate vulnerabilities across the full software and infrastructure stack. The announcement comes amid growing concern that the pace of modern software development, increasingly driven by AI coding tools, is outstripping the ability of traditional security approaches to keep systems safe.
Funding Round
The Series A round was led by Accel, with participation from Alt Capital, BoxGroup, Liquid 2 Ventures, Mantis VC, SV Angel, and a group of prominent angel investors. Among the individual backers are Jeff Dean, Kirsten Green, Colin Evans, Logan Kilpatrick, and Julian Schrittwieser, reflecting a blend of expertise across AI research, enterprise software, and venture investing. The funding brings depthfirst’s total capital raised to $40 million and signals strong investor confidence in the company’s technical direction and market opportunity.
Market Context
Software security challenges are intensifying as AI-generated code becomes more widespread across industries, increasing both speed and complexity in development cycles. At the same time, organizations are facing a new class of threats that are automated, persistent, and capable of operating at machine speed rather than human scale. depthfirst is positioning its platform as a response to this shift, aiming to give defenders AI-powered tools that can match the sophistication and velocity of AI-enabled attackers.
Product Performance
According to the company, depthfirst’s General Security Intelligence platform has shown significant gains in early benchmarks and customer deployments. In the four months since product launch, its AI agents reportedly uncovered eight times more true-positive vulnerabilities than traditional static analysis tools while reducing false positives by 85 percent. The platform also achieved state-of-the-art results on CyberGym, a widely used cybersecurity evaluation framework, delivering a 90 percent performance improvement over previous benchmarks.
Customer Adoption
Since making its product generally available, depthfirst has signed customers including Lovable, Supabase, Moveworks, and AngelList. Security leaders at these companies describe the platform as functioning like an autonomous senior security engineer that understands code context and improves over time. Customer feedback highlights faster identification of critical issues, clearer remediation guidance, and measurable gains in security team efficiency and code quality.
Team and Outlook
Founded in 2024, depthfirst was established with the goal of securing the software systems that underpin modern society, which the company views as increasingly vulnerable in an AI-driven world. Its founding team brings experience from organizations such as Google DeepMind, Databricks, and Faire, combining advanced AI research with practical security engineering. The new funding will be used to expand research and development, scale go-to-market efforts, and hire across applied research, engineering, product, and sales.
Investors see depthfirst as addressing a critical gap in a software security market that has long relied on legacy tools designed for earlier threat models. Accel partner Sara Ittelson described the company as well positioned to transform a $400 billion segment of the enterprise market as AI and agentic systems become more widely adopted. As software continues to be written faster than it can traditionally be secured, depthfirst’s approach underscores a broader shift toward AI-native defenses built for an always-on, automated threat environment.