VBT Task Author | $60 per Hour Codefeast Remote | Multiple Countries Role: VBT Task Author – AI Research & Evaluation

Experience:
3+ years

Job Type
Contractor

Location:
India Joining: Immediate Compensation: $60 per hour Average Task Time: 6 hours Role Overview We’re looking for analytical researchers to create and test challenging web-research tasks that identify weaknesses in advanced AI models. The work involves deep research, fact-checking, source verification, and evaluating AI-generated results. What You’ll Do
• Research hard-to-find facts using advanced web search and OSINT techniques.
• Verify information using reliable sources and document source URLs.
• Create clear, challenging benchmark tasks with accurate answer keys.
• Test AI models and identify errors in their research and reasoning.
• Review and improve tasks to maintain accuracy and quality. Key Requirements
• 3+ years of professional experience.




• A Bachelor’s degree or higher in a relevant engineering domain is mandatory.
• Experience in investigative research, fact-checking, AI evaluation, or technical QA.
• Familiarity with AI benchmarks such as GAIA, BrowseComp, SimpleQA, or SWE-bench.
• Strong web research, OSINT, analytical, and writing skills.
• Ability to verify sources and diagnose AI model failures. Work Commitment
• Full-time: 40 hours/week or part time: 20 hours/week.
• Four hours of overlap with PST.
• Remote contractor engagement. Selection Process
• Complete the Interest Check Form
• Complete the Google Form assessment included in the ICF.
• Profile review and onboarding for selected candidates. Interested? Apply now to contribute to cutting-edge AI research and evaluation projects with Turing

📌 VBT Task Author | $60 per Hour (India)
🏢 Codefeast
📍 India