Skip to content
Search jobs inside your AI agent. Claude, ChatGPT, Claude Code, Codex, Gemini, Cursor and Grok
Connect

Mesh Security

NewBe an early applicant

Security Researcher

Posted Updated

Description

Mesh Security is building a platform that maps and defends how modern organizations actually get attacked—across cloud, SaaS, identity, and AI integrations. We're hiring Security Researcher to drive the research behind it: tracking real-world attacker techniques, analyzing vulnerabilities and their exploitability, and ensuring our security content is accurate, current, and credible. This is a pure research role—not a software engineering position. Your job is to know how attacks work, prove it, and communicate it clearly.


What you'll do:

  • Discover new attack techniques—not just track disclosed ones. Hunt for novel attack classes across cloud, SaaS, and identity (consent-phishing chains, cross-integration token abuse, and IdP compromise paths), and take them public: CVEs credited to you, advisories, conference talks
  • Attack AI agent systems as a core charter, not a side project—break MCP servers, LLM tool chains, and agent integrations in a standing lab of real deployments, in territory new enough that you can still be first
  • Run full intrusions against live tenants (AWS, Entra ID, Okta, real SaaS sandboxes) and try to defeat our own product—every detection you evade is a gap you found, and closing it carries your name
  • Mine a dataset almost nobody else has: anonymized telemetry on how real organizations configure—and misconfigure—their cloud and SaaS estates, and publish first-of-its-kind empirical research from it
  • Lead our public analysis when a major breach or disclosure lands—rapid-response research out within hours, under your byline
  • Keep the craft sharp underneath it all: track attacker tradecraft, map it to MITRE ATT& CK, and judge which newly disclosed vulnerabilities actually matter in the real world


What we're looking for:

  • 5+ years in security research, threat intelligence, threat hunting, red teaming, or vulnerability research
  • Deep, demonstrable knowledge of attacker techniques - you can walk through how a real intrusion unfolded, step by step
  • Strong grasp of cloud/SaaS/identity attack surfaces (AWS, Azure/Entra ID, Okta, OAuth, APIs)
  • Public evidence of your work is a strong plus: CVE credits, bug bounty profile, blog posts, CTF results, or conference talks
  • Excellent written communication - your research is only as good as your ability to explain it
  • Scripting ability (e. g., Python) to support your own research is useful, but this is not a developer role


Nice to have:

  • Research or hands-on experimentation with AI/LLM security (prompt injection, attacking AI agents)
  • Certifications such as OSCP/OSEP/OSWE, GCTI, or GCFA
  • Experience presenting at security conferences (Nullcon, C0C0n, BSides, etc. )

Skills

Apply

See also

Security jobs by country — openings, pay and top skills →