Security Engineer

CodeIntegrity, Inc.; United States; Remote; USD 120,000 - 170,000; Full time

Benefits offered by this job

Competitive package

Equity included

Paid vacation

Sick leave

Parental leave

Conference budget

MacBook provided

Medical/dental/vision coverage

401(k) plan

Job summary

CodeIntegrity, Inc. is seeking a security-focused role to identify weaknesses in AI agents, tools, and data, and translate findings into safeguards. You will perform threat analysis, hands-on testing, product reviews, and collaborate with engineers to improve agent safety.

You will reproduce attacks, build detections and policies, and help measure security effectiveness across systems. This position offers hands-on impact in production AI environments and opportunities for growth.

Qualifications

• Hands-on experience finding and fixing application, API, or cloud security weaknesses.
• Ability to read production code and write tools or tests to investigate vulnerabilities.
• Understanding of authentication, authorization, trust boundaries, and sensitive data exposure.
• Ability to assess security findings and explain practical remediation to engineers.
• Willingness to learn how agents use tools and data, test assumptions, and measure safeguards.

Responsibilities

• Trace how instructions, identities, permissions, and data move between agents, tools, and servers.
• Reproduce attacks such as prompt injection, data exposure, authorization failures, and unsafe tool use with repeatable test cases.
• Build and evaluate detections, access policies, and execution isolation with software engineers.
• Measure blocked and missed attacks, as well as false positives, to assess control effectiveness.
• Review designs and code, communicate findings and remediation options, and verify fixes.
• Convert security findings into automated evaluations and regression tests, documenting control limitations.

Skills

Security testing

Code reading

Authn/Authz security

Vulnerability investigation

Job description

Role overview

Find security weaknesses in AI agents and the tools and data they use, then turn that research into practical safeguards. The work spans threat analysis, hands-on testing, product reviews, and collaboration with engineers to help organizations use agents more safely.

Responsibilities

• Trace how instructions, identities, permissions, and data move between agents, tools, and servers.
• Reproduce attacks such as prompt injection, data exposure, authorization failures, and unsafe tool use with repeatable test cases.
• Build and evaluate detections, access policies, and execution isolation with software engineers.
• Measure blocked and missed attacks, as well as false positives, to assess control effectiveness.
• Review designs and code, communicate findings and remediation options, and verify fixes.
• Convert security findings into automated evaluations and regression tests, documenting control limitations.

Requirements

• Hands-on experience finding and fixing application, API, or cloud security weaknesses.
• Ability to read production code and write tools or tests to investigate vulnerabilities.
• Understanding of authentication, authorization, trust boundaries, and sensitive data exposure.
• Ability to assess security findings and explain practical remediation to engineers.
• Willingness to learn how agents use tools and data, test assumptions, and measure safeguards.

Nice to have

• Familiarity with MCP, tool-calling agents, or LLM application security.
• Experience with prompt injection, data exfiltration, sandboxing, or adversarial testing.
• Background in ML security, detection engineering, application security, or security product engineering.

Benefits and work setup

• Salary, equity, and benefits are offered as part of a competitive package.
• Four weeks of paid vacation, paid sick leave, and paid parental leave.
• Budget for conferences, courses, and certifications; a MacBook and other needed equipment.
• Medical, dental, and vision coverage, plus a 401(k) plan.
• Direct work with organizations adopting AI in production.