Security Researcher, AI Systems

CodeIntegrity; San Francisco (CA); On-site; USD 120,000 - 180,000; Full time

Benefits offered by this job

Competitive package

Paid time off

Professional development

Top-tier equipment

Customer access

Job summary

CodeIntegrity is building the platform security layer to control MCPs and agents across enterprises. We are a small team of engineers shipping fast, backed by notable investors, focused on making enterprise AI safe to use.

We seek a security-focused researcher who can read code, prototype, and analyze how AI agents fail in real systems, delivering concrete examples and actionable guidance for engineers and product teams.

Qualifications

• Understand application security, AI systems, or both.
• Able to read code, build small prototypes, and test how systems fail.
• Capable of explaining technical risks in plain language.
• Turn ambiguous ideas into concrete examples.

Responsibilities

• Investigate how AI agents fail in real systems and convert findings into security research.
• Research agent security risks across tools, data access, identity, and MCP workflows.
• Reproduce real failure modes such as prompt injection and data exposure.
• Build test apps and demos to make risks concrete.
• Write clear technical analyses for engineers and security teams.
• Create reproducible examples to help the team understand agent failures.
• Translate findings into practical guidance and product recommendations.

Skills

Agent security research

Security testing

Code reading & prototyping

Clear technical communication

Job description

AI agents are changing how enterprises operate. Companies want to move fast with MCPs and agents, but they need a way to do it safely.

CodeIntegrity is the platform security teams use to control MCPs and agents, built for companies deploying AI across the enterprise. We raised $5 million from SYN Ventures, Antler, and Boost, and we are a small team, mostly engineers, shipping fast. We are building the infrastructure that will make enterprise AI safe to use.

What You Will Do

You'll investigate how AI agents fail in real systems and turn that work into clear security research.

• Research agent security risks across tools, data access, identity, and MCP workflows.
• Reproduce real failure modes such as prompt injection, tool misuse, data exposure, and unsafe delegation.
• Build test apps and demos that make risks concrete.
• Write clear technical analysis for engineers and security teams.
• Create reproducible examples that help the team understand real agent failure modes.
• Turn findings into practical examples, security guidance, and product recommendations.

What We Need

• You understand application security, AI systems, or both.
• You can read code, build small prototypes, and test how systems fail.
• You can explain technical risks in plain language.
• You can turn ambiguous ideas into concrete examples.
• You care about accuracy, clarity, and practical impact.

Nice to have:

• Familiarity with MCP, agents, tool calling, or LLM application security.
• Experience with prompt injection, data exfiltration, AppSec, or red-team work.
• Public writing, research notes, talks, demos, or technical blogs.

What We Offer

We offer a competitive package for people who can move quickly, work with enterprise customers, and help build critical infrastructure for enterprise AI.

• Competitive package. Salary, equity, and benefits aligned with the scope and ownership of the role.
• Time off. Four weeks of paid vacation, paid sick leave, and paid parental leave.
• Professional development. Budget for conferences, courses, and certifications across AI, enterprise software, and security.
• Top-tier equipment. Your choice of laptop and accessories for the way you work best.
• Customer access. Work directly with companies adopting AI in production and see the impact of what you build.