AI Red Team Specialist (Remote)

BLEND is building a team of security testers for an AI red-team programme with one of our clients. Instead of testing a website or app, you will test AI systems: the models, chatbots, and AI agents that follow instructions and use tools.

Your job is to make these systems misbehave, then write up how you did it. If you have a background in security testing, bug bounty, or ethical hacking and want to move into AI, this is a great fit.

What you will do

• Test AI chatbots and agents for weaknesses: tricking them past their rules, getting them to reveal hidden instructions, misusing their tools, or leaking data they should protect.
• Show real impact, not just a clever trick.
• Write clear reports: what you did, how to reproduce it, and why it matters.

What we are looking for

• Some experience in security testing, penetration testing, bug bounty, or hands-on AI/LLM hacking (work projects or personal research both count).
• Good written English, since the report is the main deliverable.
• Something to show your skills: a bug bounty profile, a CVE, a write-up, a certification, or a CTF profile.
• Willingness to pass an identity check and sign an NDA.

Nice to have (not required)

• Web penetration testing experience.
• Certifications such as OSCP, OSWE, or GIAC.
• Some Python or JavaScript.
• A Hack The Box, TryHackMe, or CTFtime profile.

How to apply

Fill in our short form (about 10 minutes). If it is a good match, you take a quick skills test, and those who pass are invited to a hands-on practical challenge.

We accept applications only via the apply link below. This is an automated self-onboarding process: https://talents.getblend.com/campaigns/ai-red-team