Company

Technology

AISafetyExperts

$0–0k Bulgaria CONTRACT Remote Friendly
Market Sentiment
HIGH DEMAND

Neural analysis suggests this role is
optimal for Mid+ candidates.

The Brief

“AI Safety Experts. Skills: AI Safety, Red Teaming, Adversarial Testing. Design adversarial test cases. Execute adversarial test cases”

What You'll Achieve.

Contribute directly to AI safety research; Contribute directly to AI alignment research

Industry & Context.

Technology
Problems you'll solve

Identify system weaknesses; Identify edge-case behaviors

Eligibility Requirements

Remote work, Independent scheduling

What They're Looking For.

Must Have

Fluency in English and Odia, Prior experience in AI red teaming, Prior experience in cybersecurity, Prior experience in adversarial testing, Prior experience in socio-technical risk analysis, Ability to think creatively, Ability to think critically, Structured thinking skills, Ability to follow frameworks, Ability to document findings clearly, Excellent written communication skills, Ability to work independently

Nice to Have

Familiarity with AI systems, Familiarity with conversational agents, Familiarity with machine learning concepts

What You'll Do.

Design adversarial test cases

Execute adversarial test cases

Evaluate AI model behavior

Identify vulnerabilities

Classify vulnerabilities

Document vulnerabilities

Identify risk patterns

Classify risk patterns

Document risk patterns

Generate structured datasets

Generate annotated examples

Follow defined taxonomies

Follow safety frameworks

Ensure consistent evaluation standards

Produce clear reports

Produce reproducible reports

Describe observed risks

Describe system weaknesses

Collaborate on improving evaluation coverage

Expand test scenarios

How You'll Work.

Team & Collaboration

Collaborate on improving evaluation coverage

Communication Scope

Written communication

Full Job Description

## Accountabilities Design and execute adversarial test cases to evaluate AI model behavior, including jailbreaks, prompt injections, and multi-turn manipulation scenarios. Identify, classify, and document vulnerabilities, failures, and risk patterns in AI-generated outputs. Generate structured datasets and annotated examples that support AI safety evaluation and training. Follow defined taxonomies, benchmarks, and safety frameworks to ensure consistent evaluation standards. Produce clear, reproducible reports describing observed risks and system weaknesses. Collaborate on improving evaluation coverage by expanding test scenarios across linguistic and cultural contexts (English and Odia). Requirements: Strong fluency in both English and Odia (native or near-native proficiency required). Prior experience in AI red teaming, cybersecurity, adversarial testing, or socio-technical risk analysis. Ability to think creatively and critically to identify system weaknesses and edge-case behaviors. Strong structured thinking skills, with the ability to follow frameworks and document findings clearly. Excellent written communication skills for both technical and non-technical audiences. Familiarity with AI systems, conversational agents, or machine learning concepts is a plus. Ability to work independently in a remote, task-driven environment. Benefits: Competitive hourly compensation of 20 to 22 USD per hour. Fully remote and flexible work arrangement with independent scheduling. Opportunity to contribute directly to cutting-edge AI safety and alignment research. Exposure to advanced red-teaming methodologies and frontier AI systems. Weekly payments via Stripe or Wise as an independent contractor. Ongoing project extensions based on performance and impact. ## Additional Information How Jobgether works: We use an AI-powered matching process to ensure your application is reviewed quickly, objectively, and fairly against the role's core requirements. Our system identifies the top-

Free ATS check

Applying for this AI Safety Experts role?

Most applicants get filtered before a human reads their resume. See if yours makes the cut.

How to Apply on Lever

  • Lever uses a streamlined one-page form — apply in under 5 minutes.
  • LinkedIn import works well; review parsed data before submitting.
  • The cover letter field is optional but visible to reviewers — use it to differentiate.
  • Referral codes from employees can significantly boost visibility of your application.

ANONYMOUS · UNFILTERED

What do employees actually say about this company?

Real rants from real employees. Read before you apply.

Read Company Rants →