Senior Security Researcher, Voice Agent Red Team

present of AI; San Francisco (CA); On-site; USD 180,000 - 240,000; Full time

Benefits offered by this job

Live environment testing

Clear security reports

Hands-on AI attack practice

Job summary

Hamming AI is seeking a Senior Security Researcher to lead security assessments of production voice and chat agents in finance and health sectors. The role requires five years of offensive security experience, hands-on AI attack skills including prompt injection and tool abuse, and the ability to deliver clear, reproducible findings.

You will analyze how models handle identity checks, tool calls, and retrieved content, and translate findings into actionable steps for developers.

Qualifications

• Five+ years of offensive security experience on production systems.
• Hands-on AI attack skills such as prompt injection and tool abuse.
• Ability to write clear, reproducible findings for developers.

Responsibilities

• Conduct authorized security assessments of production voice and chat agents.
• Evaluate how models handle identity checks, tool calls, and retrieved content.
• Produce structured, actionable security reports for engineering teams.

Skills

Offensive security

Prompt injection

Tool abuse

Reproducible reports

Tools

Python

TypeScript

Job description

Senior Security Researcher, Voice Agent Red Team

This role requires running authorized security assessments of production voice and chat agents for banks and health systems. A strong candidate brings at least five years of offensive security experience, hands-on AI attack skills like prompt injection and tool abuse, and the ability to write clear, reproducible findings. They must also work carefully in live environments and verify that fixes actually hold.

How it fits the AI era

AI changes the work by making conversational agents a new attack surface, so the researcher must test how models handle identity checks, tool calls, and retrieved content. The human part stays the same: scoping engagements, judging which failures have real consequences, and writing findings that developers can act on. AI tools help move faster, but the judgment and evidence remain central.

How it builds AI experience

The role offers direct experience attacking production voice and chat agents, including prompt injection, jailbreak chains, and data exfiltration through model outputs. Building this experience means practicing authorized assessments on agent systems and learning to turn manual findings into repeatable tests. Working at Hamming, which focuses on voice AI security, provides exposure to real enterprise deployments and responsible disclosure.

Worth showing

• Offensive security assessments
• Tool call abuse
• Reproducible security reports
• Python or TypeScript

This is our own read, written from the employer's posting and the company's Spotlight. The employer's posting is the source of truth for the role.

About Hamming AI

Complete QA testing, red-teaming, and production observability platform for conversational voice AI agents, per the company.