Skip to main content
Loading
loadingbar
Loading, Please wait..!!

Ai Safety Expert Adversarial Ml

  • Ai Safety Expert Adversarial Ml job type Posted on: Nov 19, 2026
  • Ai Safety Expert Adversarial Ml experience level Mercor
  • Ai Safety Expert Adversarial Ml employment type San Francisco, California
  • Ai Safety Expert Adversarial Ml salary Salary: $16 per hour
  • Ai Safety Expert Adversarial Ml employment type Remote
  • Ai Safety Expert Adversarial Ml salary Full-time

Point Apply Here APPLY LATER

Curious about compensation?

Explore the historical salary trends, average pay, and estimated compensation for Ai Safety Expert Adversarial Ml roles in California.

View Salary Guide →

Job Title :

Ai Safety Expert Adversarial Ml

Job Type :

Full-Time

Job Location :

San Francisco California United States

Remote :

Yes

Jobcon Logo Job Description :

Job DescriptionJob DescriptionAbout the job\nMercor connects elite creative and technical talent with leading AI research labs. Headquartered in San Francisco, our investors include Benchmark, General Catalyst, Peter Thiel, Adam D'Angelo, Larry Summers, and Jack Dorsey.\nPosition: AI Safety Experts — English & UrduType: ContractCompensation: $16–$22/hourLocation: Remote\nRole Responsibilities\n\n Red team conversational AI models and agents to identify jailbreaks, prompt injections, and misuse cases.\n Generate high-quality human data by annotating failures, classifying vulnerabilities, and flagging systemic risks.\n Apply structure by following taxonomies, benchmarks, and playbooks to ensure consistent testing.\n Document reproducibly to produce reports, datasets, and attack cases that customers can act on.\n Work independently and asynchronously to meet deadlines while improving AI model performance.\n\nQualifications\nMust-Have\n\n Native fluency in English and Urdu.\n Strong judgment about language and content accuracy.\n Rigorous attention to detail and ability to notice subtle errors.\n Ability to work to guidelines and quality standards consistently.\n Strong communication skills for both technical and non-technical audiences.\n Adaptability to move across projects, task types, and customers.\n\nPreferred\n\n Experience in Adversarial ML: jailbreak datasets, prompt injection, RLHF/DPO attacks, model extraction.\n Background in Cybersecurity: penetration testing, exploit development, reverse engineering.\n Knowledge of socio-technical risk: harassment/disinfo probing, abuse analysis, conversational AI testing.\n Creative probing skills: psychology, acting, writing for unconventional adversarial thinking.\n\nApplication Process (Takes 20–30 mins to complete)\n\n Upload resume\n AI interview based on your resume\n Submit form\n\nResources & Support\n\n For details about the interview process and platform information, please check: https://talent.docs.mercor.com/welcome\n For any help or support, reach out to: support@mercor.com\n\nPS: Our team reviews applications daily. Please complete your AI interview and application steps to be considered for this opportunity.

Jobcon Logo Position Details

Posted:

Nov 19, 2026

Reference Number:

16275_e972d3111

Employment:

Full-time

Salary:

$16 per hour

City:

San Francisco

Job Origin:

APPCAST_CPC

Share this job:

  • linkedin

Jobcon Logo
A job sourcing event
In Dallas Fort Worth
Aug 19, 2017 9am-6pm
All job seekers welcome!

Ai Safety Expert Adversarial Ml    Apply

Click on the below icons to share this job to Linkedin, Twitter!

Job DescriptionJob DescriptionAbout the job\nMercor connects elite creative and technical talent with leading AI research labs. Headquartered in San Francisco, our investors include Benchmark, General Catalyst, Peter Thiel, Adam D'Angelo, Larry Summers, and Jack Dorsey.\nPosition: AI Safety Experts — English & UrduType: ContractCompensation: $16–$22/hourLocation: Remote\nRole Responsibilities\n\n Red team conversational AI models and agents to identify jailbreaks, prompt injections, and misuse cases.\n Generate high-quality human data by annotating failures, classifying vulnerabilities, and flagging systemic risks.\n Apply structure by following taxonomies, benchmarks, and playbooks to ensure consistent testing.\n Document reproducibly to produce reports, datasets, and attack cases that customers can act on.\n Work independently and asynchronously to meet deadlines while improving AI model performance.\n\nQualifications\nMust-Have\n\n Native fluency in English and Urdu.\n Strong judgment about language and content accuracy.\n Rigorous attention to detail and ability to notice subtle errors.\n Ability to work to guidelines and quality standards consistently.\n Strong communication skills for both technical and non-technical audiences.\n Adaptability to move across projects, task types, and customers.\n\nPreferred\n\n Experience in Adversarial ML: jailbreak datasets, prompt injection, RLHF/DPO attacks, model extraction.\n Background in Cybersecurity: penetration testing, exploit development, reverse engineering.\n Knowledge of socio-technical risk: harassment/disinfo probing, abuse analysis, conversational AI testing.\n Creative probing skills: psychology, acting, writing for unconventional adversarial thinking.\n\nApplication Process (Takes 20–30 mins to complete)\n\n Upload resume\n AI interview based on your resume\n Submit form\n\nResources & Support\n\n For details about the interview process and platform information, please check: https://talent.docs.mercor.com/welcome\n For any help or support, reach out to: support@mercor.com\n\nPS: Our team reviews applications daily. Please complete your AI interview and application steps to be considered for this opportunity.

Loading
Please wait..!!