depthfirst
Category: AI in Cybersecurity
An applied AI lab building 'General Security Intelligence' through autonomous agents designed to identify, analyze, and remediate software vulnerabilities at scale. depthfirst was founded in 2024. The company is led by Qasim Mithani. Based in San Francisco, USA. Team size: 11-50. Total funding raised: $40M. Latest round: Series A ($40M, Jan 2026). Key investors include Accel, Alt Capital, BoxGroup, Liquid 2 Ventures, Mantis VC, SV Angel, Jeff Dean, Kirsten Green.
- Founded
- 2024
- Headquarters
- San Francisco, USA
- Team size
- 11-50
- Total funding
- $40M
Value proposition
Automates the end-to-end security lifecycle—from discovery to remediation—using AI agents that outperform traditional security scanners by 90% on industry benchmarks, significantly reducing manual security debt.
Products and solutions
Autonomous Vulnerability Research Agents, Automated Remediation & Patching System, General Security Intelligence (GSI) Engine, AI-Native Security Benchmarking Tools
Unique value
Moves beyond traditional 'static analysis' by utilizing agentic AI that mimics the reasoning of human security researchers to not only find bugs but autonomously generate fixes.
Target customer
Enterprise software companies, DevSecOps teams, and organizations managing large-scale cloud infrastructure and complex codebases.
Industries served
Cybersecurity, Cloud Infrastructure, Software Development (DevSecOps), Enterprise SaaS
Technology advantage
Proprietary AI models optimized for security logic that achieve a 90% performance lead over legacy scanners on CyberGym benchmarks; founded by a 'super-team' of experts from DeepMind, Databricks, and AWS.
How they differentiate
Utilizes 'General Security Intelligence' (GSI) and agentic AI that mimics human reasoning to not only detect but autonomously remediate vulnerabilities, outperforming traditional scanners by 90% on CyberGym benchmarks.
Main competitors
Snyk, Wiz, Cognition AI (Devin)
Key partnerships
Accel (Lead Series A Investor), Strategic investors from Databricks, Google DeepMind, and Faire, CyberGym (Benchmark validation partner)
Notable customers
Stealth Enterprise Software Clients, DevSecOps Teams, Cloud Infrastructure Providers
Major milestones
Founded in 2024 by engineering leaders from Databricks, Google DeepMind, and AWS, Developed proprietary 'General Security Intelligence' engine for autonomous vulnerability research, Closed $40M Series A led by Accel in January 2026
Growth metrics
Achieved a 90% performance lead over legacy security scanners on industry benchmarks; scaled team to 11-50 employees within two years of founding.
Market positioning
AI-native 'Agentic Security' pioneer focusing on autonomous vulnerability research and automated patching.
Geographic focus
North America (Headquartered in San Francisco), with a global focus on enterprise cloud infrastructure.
Patents and IP
No registered patents publicly disclosed as of January 2026; focus is on proprietary model architectures and agentic workflows.
About Qasim Mithani
Qasim Mithani is a seasoned engineering leader with a decade of experience scaling critical infrastructure at top-tier tech companies. Before founding depthfirst, he spent 5 years at Databricks as a Senior Engineering Leader, where he was instrumental in scaling core engineering during their hyper-growth phase. Prior to that, he spent 4 years at Amazon Web Services (AWS) as a Senior Engineering Manager focused on secure cloud services and infrastructure.
Official website: https://depthfirst.com/