Security Engineer

Agentplace; United States; Remote; USD 51,000 - 61,000; Part time

Benefits offered by this job

Remote contract

Job summary

Agentplace is seeking a security-focused role to protect customer-facing AI agents and the data they access. You will assess abuse vectors in agents, integrations, and access controls, and implement protections with repeatable tests.

You will build threat models, test for prompt injection and data leakage, design data isolation, and fix vulnerabilities with verifiable tests. The role is a remote contract with flexible hours, starting at $5,000 per month.

Qualifications

• Experience in application security or product security.
• Understanding of risks specific to large language model systems, including prompt injection and data exfiltration.
• Ability to write code for security tests and remediation.
• Clear communication of vulnerabilities, impact, and risk severity.

Responsibilities

• Build threat models for agents, their integrations, and the information those systems handle.
• Test for prompt injection, data leakage, and attempts to trigger unauthorized actions.
• Design access controls that keep each customer’s data isolated.
• Implement fixes for discovered vulnerabilities and verify them with repeatable tests.
• Document security risks, explain their severity, and write practical guidance for builders and customers.

Skills

Application security

LLM risk awareness

Security testing code

Communication of vulnerabilities

Job description

Role overview

Secure customer-facing AI agents and the data and services they can access. You’ll assess how agents, integrations, and access controls could be abused, then implement protections and demonstrate through repeatable tests that the risks have been reduced.

Responsibilities

• Build threat models for agents, their integrations, and the information those systems handle.
• Test for prompt injection, data leakage, and attempts to trigger unauthorized actions.
• Design access controls that keep each customer’s data isolated.
• Implement fixes for discovered vulnerabilities and verify them with repeatable tests.
• Document security risks, explain their severity, and write practical guidance for builders and customers.

Requirements

• Experience in application security or product security.
• Understanding of risks specific to large language model systems, including prompt injection and data exfiltration.
• Ability to write code for security tests and remediation.
• Clear communication of vulnerabilities, impact, and risk severity.
• Relevant experience may include penetration testing, AI red teaming, or security engineering for SaaS products.

Benefits and work setup

Remote contract engagement with flexible hours. Compensation starts at $5,000 per month. Freelancers and solo agencies may continue serving their own clients.