Listen to this articledepthfirst secured an $80 million Series B round, led by Meritech Capital, bringing its total funding to $120 million just months after its Series A, to accelerate development of specialized AI security models and expand enterprise adoption of its General Security Intelligence platform.
depthfirst’s latest funding is an $80 million Series B round. Led by Meritech Capital with participation from Forerunner Ventures and The House Fund, alongside returning investors Accel, Box Group, Liquid 2 Ventures, Alt Capital, and Mantis VC, the round brings the company’s total capital raised to $120 million following its $40 million Series A just under 90 days earlier. This rapid follow-on reflects exceptional investor conviction in depthfirst’s execution speed and its positioning at the intersection of applied AI and software security.
How will depthfirst use the funds?
The funding will fuel three core priorities: training additional domain specific security models across new areas, expanding the AI research team, and accelerating enterprise adoption. This allocation directly supports depthfirst’s strategy of owning both the product experience and the underlying intelligence in an AI native security platform called General Security Intelligence (GSI). GSI analyzes code, infrastructure, and business logic end to end to detect complex vulnerabilities, prioritize real risks, reduce alert noise dramatically, and deliver “ready to merge” fixes directly in developer workflows.
A major milestone tied to the announcement is the launch of dfs-mini1, depthfirst’s first in-house security model. Initially focused on securing cryptocurrency smart contracts, dfs-mini1 was built by post training an open source base model through reinforcement learning in security specific environments. It was evaluated on OpenAI’s EVMBench benchmark for smart contract vulnerabilities. Early results show dfs-mini1 outperforming frontier general purpose models while running at 10x to 30x lower inference cost. Internal evaluations further indicate strong generalization to other security tasks, validating the transferability of its domain specialized training approach. As CTO Andrea Michi noted, owning the training process allows optimization for what matters most, vulnerability detection and verification, yielding a model that is cheaper, more accurate, and more responsive to ongoing investment than generic alternatives.
This technological edge comes at a critical time. AI has accelerated software development, enabling unprecedented velocity in code releases, services, and infrastructure changes. Simultaneously, attackers leverage AI to discover and exploit weaknesses at scale, overwhelming legacy security tools never designed for this pace. depthfirst’s platform counters this by deploying autonomous agents that reason across entire systems 24/7, achieving orders of magnitude higher recall with a fraction of the noise of traditional approaches. Since general availability in late 2025, the company has secured partnerships with Fortune 500 enterprises and high growth software firms including ClickUp, Lovable, Supabase, incident.io, and Moveworks. Developer adoption metrics are strong: approximately 80% of its fix recommendations are accepted and merged, with reported reductions in alert noise by 8x and false positives by 85%, alongside a 70% cut in security engineering workload for some customers.
Recommended: Syndigo Acquires Taggstar
The investor syndicate underscores the thesis. Meritech’s Arsham Memarzadeh highlighted the “rare combination of research capabilities and security experience” that enables depthfirst to deliver precisely what customers need amid sophisticated AI enabled attacks. Accel’s Sara Ittelson emphasized the company’s dual ownership of product and underlying intelligence, positioning it to fundamentally reshape how modern systems are secured. CEO Qasim Mithani framed the moment as one where public market signals show investors recognizing AI’s disruption of the legacy security stack, but stressed that winning requires security specific models embedded in optimized workflows, capabilities depthfirst’s team, with its DeepMind, AWS, Databricks, Cash App, and Faire pedigrees, is uniquely equipped to build.
The Series B cements depthfirst’s trajectory as a leader in AI driven software security. In under three months since emerging from stealth, the company has moved from initial product availability to meaningful customer traction, proprietary model development, and a substantial capital base. The $120 million total positions it to scale research, product, and go to market aggressively in a market where AI both amplifies threats and offers the only viable defense. By prioritizing specialized models over general purpose tools, depthfirst is building durable competitive moats that align product delivery with real world security outcomes, setting the stage for broader domain expansion and enterprise dominance.
Please email us your feedback and news tips at hello(at)techcompanynews.com
