Chemical Safety AI Model Evaluator

OpenTrain AI; Vancouver; Hybrid; CAD 96,000 - 138,000; Part time

Benefits offered by this job

Remote work

Hourly contract

Contractor, part-time

Job summary

OpenTrain AI is seeking a Chemical Safety AI Model Evaluator to assess AI responses to chemistry-related questions. The role is remote, part-time, and paid hourly, with contractor engagement. Requires relevant expertise or background and English language work for 20+ hours per week.

You will review prompts and model answers, ensuring safety, accuracy, and contextual appropriateness, and you will craft corrected answers and training prompts to improve model behavior.

Qualifications

• Ability to distinguish benign, dual-use, and adversarial questions with sound safety judgment.
• Ability to evaluate technical answers for accuracy, risk, and context, with clear written explanations.
• Experience with safety data sheets, exposure assessment, incident reports, or regulatory review is valuable.

Responsibilities

• Review realistic chemical safety requests and multiple model answers to judge safety, accuracy and appropriateness.
• Create prompts and responses for chemical safety model training.
• Explain reasoning behind judgments and identify why a dual-use request is sensitive.
• Rank model responses from strongest to weakest and write the corrected answer.

Skills

Chemistry safety evaluation

Safety judgment & writing

Technical reasoning & explanation

Clear written communication

Regulatory familiarity

Education

Relevant degree or professional background

Job description

About OpenTrain

OpenTrain AI contracts specialists to help build safer, more capable artificial intelligence. OpenTrain is the platform where people discover AI training work, build a professional profile, and grow a lasting portfolio in this fast-growing field.

• Remote contract and part-time opportunity
• Build experience in specialized AI training and evaluation

About AI Training Work

AI training is the human side of developing modern models. Experts review prompts and answers, identify errors and unsafe behavior, and provide examples that help AI systems respond more accurately and responsibly.

In this role, your chemical safety expertise will help shape how models handle sensitive technical questions, including requests with legitimate, dual-use, or adversarial purposes.

• Contribute directly to the development of advanced AI systems
• Use professional judgment to improve safety and reliability
• Work remotely with flexible part-time scheduling

The Chemical Safety AI Model Evaluator Role

OpenTrain is seeking a Chemical Safety AI Model Evaluator to assess how artificial intelligence systems respond to sensitive questions involving chemistry, toxicology, industrial hygiene, process safety, and chemical security.

The listing identifies this as an entry-level opportunity, but the work requires relevant professional expertise or a comparable subject-matter background. The expected commitment is 20 or more hours per week.

• Pay: $70-$100 per hour
• Payment type: Hourly contract
• Work arrangement: Remote
• Engagement: Contractor, part time
• Language: English
• Data type: Text

What You’ll Do

You will review realistic chemical safety requests and multiple AI-generated answers. Your evaluations will help determine whether model behavior is accurate, safe, appropriate, and suitable for the context.

You will also create detailed prompts and responses that support learning across diverse chemical safety topics. The work includes testing models for inaccuracies, bias, unsafe behavior, and inappropriate refusals.

• Classify requests as benign, dual-use, or adversarial
• Evaluate three model answers for safety, accuracy, and appropriateness
• Explain the reasoning behind each judgment
• Rank model responses from strongest to weakest
• Write the answer the model should have provided
• Identify why a dual-use request is sensitive
• Explain the legitimate purpose and appropriate access boundaries for sensitive information
• Create prompts and responses for chemical safety model training

Required Expertise and Skills

You should be able to distinguish benign, dual-use, and adversarial chemical questions using sound safety judgment. You must also be able to evaluate technical answers for accuracy, risk, context, and appropriate boundaries, then explain your decisions clearly in writing.

A relevant degree or comparable professional background is typically expected. Experience with safety data sheets, exposure assessment, incident reports, or regulatory review is valuable.

• Expertise in synthetic or analytical chemistry, toxicology, industrial hygiene, process safety, chemical security, emergency response, or chemical regulation
• Ability to assess safety, accuracy, and risk in technical chemical information
• Ability to explain model-response judgments and write an appropriate corrected answer
• Strong written communication and clear technical reasoning
• Familiarity with safety data sheets, exposure assessment, incident reports, or regulatory review
• Relevant experience connected with EPA, OSHA, HSE, OPCW, hazmat operations, poison control, or chemical incident response is helpful

Eligibility and Availability

The role description highlights availability to citizens or permanent residents of the United States, United Kingdom, Canada, Australia, and New Zealand. The structured opportunity listing also includes Austria, Belgium, Bulgaria, Croatia, Cyprus, Czechia, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, the Netherlands, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, and Sweden.

Confirm your eligibility and location during the application process. English is the required working language, and the expected workload is 20 or more hours per week.

• Eligible locations are specified in the opportunity listing
• English-language work
• 20+ hours per week

Build Your AI Training Career

OpenTrain helps specialists manage AI training opportunities, demonstrate credible work, and develop a professional portfolio. Your evaluations can give you practical experience at the intersection of chemical safety and cutting-edge AI development.

• Showcase specialized evaluation experience in your profile
• Grow a durable portfolio in AI training and data labeling