Agent Security Research Engineer (AI)

Location: Taipei, Taiwan

Company

hirify.global Security develops a cybersecurity platform for protecting SaaS and non-human identities, applications, integrations, and AI agents.

What you will do

• Research how coding assistants, desktop agents, workflow automation tools, and MCP-connected tools can misbehave or be compromised.
• Design detection logic and security policies for agent-related threats.
• Build detections in security engines and data pipelines, test them, document them, and ship them to production.
• Create reproducible positive and negative test cases, test harnesses, and event-log parsing workflows.
• Help customers adopt security content and tune detections based on real-world results.
• Explain agent attack chains and practical risk to engineering, product, and customer-facing teams.

Requirements

• Hands-on knowledge of agentic tools, including tool calling, hooks, lifecycle events, MCP servers and transports, skills, plugins, and permission modes.
• Strong understanding of direct and indirect prompt injection, tool and description poisoning, confused-deputy issues, excessive agency, secret leakage, and malicious or over-permissioned MCP servers and extensions.
• Strong SQL skills, preferably with analytical stores such as ClickHouse, Databricks, or Snowflake.
• Proficiency in Python or Go for building test harnesses, parsing event logs, and implementing detections.
• Working knowledge of SaaS and cloud identity, including OAuth, PATs, API tokens, scopes, Git, GitHub/GitLab, and CI/CD.
• Ability to distinguish practical security risks from theoretical risks and communicate them clearly to customers.

Nice to have

• Published research, CVEs, talks, or blog posts on LLM or agent security.
• Threat detection experience and familiarity with endpoint security on macOS, Linux, or Windows.
• Experience with dbt, Dagster, or other data-pipeline tools.
• Background in SaaS security, CASB/SSPM, or insider-threat products.
• Extensive use of Claude Code, Copilot, Cursor, or similar tools and informed opinions about their security models.

Culture & Benefits

• Work on agentic AI security and protection of SaaS and non-human identity layers.
• Competitive compensation with equity and 401k for US-based employees.
• Healthcare coverage with dental and vision benefits for US-based employees.
• Flexible paid time off, paid holidays, and 12 weeks of new parent or family leave for US-based employees.
• Personal and professional development resources.