Agent Security Research Engineer

Obsidian Security is a global leader in cyber security protecting the SaaS and non-human identity layer modern enterprises run on — from Salesforce and Snowflake to the AI agents now deployed inside them. The Obsidian Security platform gives unified visibility into every human and machine identity, app, and integration across their SaaS estate, detects identity-based and supply chain attacks in real time, and enforces least-privilege access before it's exploited. Trusted by major Fortune 500 companies T-Mobile, Workday, Snowflake, and S&P Global, Obsidian ranked No. 95 on the 2025 Deloitte Technology Fast 500™, growing nearly 1000% from 2021–2024 — helping CISOs turn an unmonitored attack surface into one they can govern with confidence. Obsidian has also been recognized by Forbes as America's Best Startup Employers in 2026.

In August 2026, Obsidian raised $85 million in Series D financing led by Crescent Cove Advisors, with participation from existing investors including Greylock Partners, Menlo Ventures, Norwest Venture Partners, IVP, Wing Ventures, and GV — pushing Obsidian's valuation to $1.1 billion, crossing into unicorn status. Obsidian is using the funding to deepen its R&D in agentic AI security and extend its platform as the standard for governing what AI agents can access and do inside third-party applications.

With global momentum, a growing partner ecosystem including SentinelOne, Databricks, and Google Cloud, and fresh capital from its Series D behind it, Obsidian is scaling rapidly toward long-term growth and IPO readiness.

Agent Security Research Engineer – Taiwan

About the role: You'll own agent security content from idea to production. You'll research how AI agents (coding assistants, desktop agents, workflow automation tools, and MCP-connected tools) can misbehave or be compromised. Then you'll design the detection or policy, build it in our engines and pipelines, test it, ship it, document it, help customers adopt it, and tune it based on real-world results. You're accountable for each piece of content delivering security value in customers' environments.

Required Skills: 

• Hands-on knowledge of how agentic tools work: tool calling, hooks and lifecycle events, MCP servers and transports, skills and plugins, permission modes
• Solid grasp of agent-specific attack classes: direct and indirect prompt injection, tool and description poisoning, confused-deputy and excessive-agency issues, secret leakage into context, and malicious or over-permissioned MCP servers and extensions.
• Able to tell real risk from theoretical risk and explain the difference to a customer.
• Strong SQL, ideally on analytical stores like ClickHouse, Databricks, or Snowflake.