An enterprise client helps the world’s most innovative companies improve their AI models by providing human feedback.
Job DescriptionThe client is seeking a high volume of freelancers to have a conversation with two AI models, and then to write a comparison and evaluation of the two model performances.
What You'll Do:
- Click “start”, talk to Model A, then Model B, out loud, from your device
- Using the provided scenario and guidelines, start a dialogue with each model
- Have a natural conversation for 2-6 turns
- After completing both conversations, compare and rank the models based on provided criteria
- Submit your Pass/Partial/Fail vote for each model
For work on this project, you may be asked to write questions/prompts and an evaluation the AI model’s answer, and/or contribute a variety of video and audio-based tasks. Some projects may involve acting short scripts in different tones or emotional styles.
- Native Bengali proficiency
- College-level or higher writing experience
- Strong reasoning and analytical skills
- A microphone (a laptop microphone is sufficient)
- Physical location in an Asian country
- Fully remote opportunity
- Flexible work schedule
Skills Required
- Native Bengali proficiency
- College-level or higher writing experience
- Strong reasoning and analytical skills
- Access to a microphone
- Physical location in an Asian country
What We Do
Lifted, an Upwork Company, is a B2B SaaS platform that helps enterprise organizations source, contract, manage, and pay contingent talent globally and compliantly. It supports multiple engagement models, including independent contractors, staff augmentation, and employer-of-record, while integrating with MSPs and VMSs to provide centralized visibility, spend control, and audit-ready compliance, delivering a white-labeled talent experience for hiring managers and contingent workforce programs.
