About the role
RustLabs is hiring security engineers to test and harden AI language models. You will attack LLMs the way a real adversary would, document what breaks, and help build the data that makes these systems safer.
What you will do
Red-team LLMs and AI agents: prompt injection, jailbreaks, data leakage, tool misuse and unsafe outputs
Design attack scenarios and write clear, reproducible reports with severity and fixes
Review model responses against security rubrics
You need
3+ years in application security, penetration testing, secure code review or AI/ML security
Hands-on experience testing LLMs or AI applications (red teaming, prompt injection, OWASP LLM Top 10)
Solid coding in Python or another language
Excellent written English and a LinkedIn profile
About 20 to 40 hours a week, starting this week
Pay and terms
About USD 22 per hour, fully remote, independent contract
How to apply
Click Apply and fill in our 2-minute form.
Originally posted on Himalayas
Before you start
Sorted before day one
Some links here earn us a commission. It costs you nothing and does not affect what we recommend.
- Health cover
An employer's health plan may not cover you abroad. Check you are covered where you will actually be.See details for SafetyWing Nomad InsuranceAffiliate link (opens in a new tab)
- Staying connected
A data plan for the countries you will work from, set up before you land.See details for AiraloAffiliate link (opens in a new tab)
- Working securely
Working on cafe or coworking Wi-Fi? A VPN keeps company logins private.See details for Proton VPNAffiliate link (opens in a new tab)