BLEND is building a pool of AI red-team specialists for a client working on generative AI models and agentic systems. The work is adversarial testing: probing models and agent workflows for prompt injection, jailbreaks, unsafe tool use, data leakage and guardrail failures, then documenting what you found and how you found it. If you already completed and passed the practical test on the external platform through our earlier “AI Red-Team” posting, there's no need to apply again — you're already in the client's pool.
The practical test is paid: you receive $25 once your report is approved by the client.
How the process runs:
You complete a short questionnaire about your security background, around 10 minutes.
If your answers meet the requirements, you go straight on to a brief technical screener on this platform.
Candidates who pass are invited to a practical test on an external security platform, where they register themselves, work through the challenge and submit a written report.
Successful candidates complete identity verification and onboarding into the pool.
What to know before you enroll:
This is a pilot. Passing the practical test places you in the client's pool but does not guarantee assigned work or a set volume of hours.
The practical test is marked by the client. Their decision is final, and in most cases we will not be able to tell you why an individual submission was rejected.
Written and spoken English is required — all briefs, reports and communication are in English.
We expect responsiveness. Candidates who do not reply within the reminder window lose their place.
