Senior AI/LLM Red Team Operator

Armadin; Palo Alto (CA); Remote; USD 180,000 - 240,000; Full time

Benefits offered by this job

Meaningful Equity Ownership

In-Office Meals

401(k), HSA, and FSA Plans

Flexible PTO

Full Health, Dental, & Vision Coverage

Job summary

Armadin is seeking a Senior Red Team Operator to lead hands‑on AI offense, test kill chains, and reveal exploitable weaknesses across AI models and pipelines. You’ll craft reproducible exploits, map impact, and guide safety through automated, multi‑agent attacks.

You’ll work with a small, autonomous team of researchers and engineers, traveling up to 20% for engagements in the United States.

Qualifications

• Deep, hands-on AI offense.
• Application security fundamentals.
• An attacker's toolkit and the instinct to go beyond it.
• Systems understanding that makes chaining possible.
• Comfort everywhere you land.
• An AI-forward way of working.
• Sharp human-in-the-loop judgment.
• A builder's, startup mindset.
• Clear communication with clients.
• Logistics: eligible to work in the United States without sponsorship; travel up to 20%.

Responsibilities

• Run full kill chains, not checklists. Attack LLMs, agents, RAG pipelines, and the applications wrapped around them the way a real adversary would.
• Prove what's exploitable. Turn jailbreaks or model behaviours into reproducible exploits with clear impact.
• Map the blast radius from foothold to data access across identity, cloud, and applications.
• Hijack the agent. Abuse tool calling, memory poisoning, and multi‑agent handoffs to edge cases.
• Poison the pipeline. Attack training data, embeddings, and MLOps infra.
• Read code like an attacker. Review frameworks and model-serving code for injection paths.
• Take models apart. Use probing and extraction to understand leaks.
• Train the swarm. Translate attack tradecraft into agent behavior at speed.
• Break the machine. Stress-test models and maintain safety in the loop.
• Build scripts and tools to automate AI attacks at scale.
• Tell the attack story with debriefs for engineers, executives, and counsel.
• Stay ahead of the threat by researching new AI attack classes.

Skills

AI offense

Agent frameworks

Web & API security

Python scripting

Cross-platform

Tools

Burp Suite

Caido

Garak

PyRIT

promptfoo

Adversarial Robustness Toolbox

Job description

Armadin is seeking a Senior Red Team Operator to lead hands‑on AI offense, test kill chains, and reveal exploitable weaknesses across AI models and pipelines. You’ll craft reproducible exploits, map impact, and guide safety through automated, multi‑agent attacks.

You’ll work with a small, autonomous team of researchers and engineers, traveling up to 20% for engagements in the United States.