AI Trainer (Dogri linguist)

An Enterprise client is seeking freelancers with Linguistics expertise in Dogri to create and refine content for generative AI models to ensure accuracy and relevance across a wide array of topics.

Labelbox is the leading AI platform for building intelligent applications. Whether teams are building AI products by using LLMs that require human fine-tuning, or applying AI to reduce the time associated with manually-intensive tasks like data labeling or finding business insights, Labelbox enables them to do so effectively and quickly. Alignerr, which is operated by Labelbox, is a community of subject matter experts from several disciplines who align AI models by creating high-quality data in their field of expertise to build the future of Generative AI.

Tasks may include:

• Educating AI: Analyze and provide feedback on AI-generated outputs in order to directly improve the AI's accuracy and ability to apply its knowledge to real-world problems
• Problem Solving: Provide step-by-step solutions and explanations to complex problems to teach the AI how to reason through these problems effectively
• Red Teaming: Identify potential biases, limitations, or inaccuracies in the AI's knowledge base. Design and conduct tests that push the boundaries of the model's understanding, ensuring its outputs are reliable and applicable to real-world scenarios.

• Native Dogri language proficiency with excellent writing skills is required
• Professional writing experience as a researcher, journalist, technical writer, editor, or similar is preferred but not required
• Experience with similar AI-training tasks is preferred but not required

• The rate for this project is $15 per hour.
• The volume of work may fluctuate throughout the duration of the contract, with an anticipated need for 10-40 hours of work per week.
• A lump sum milestone payment will be released every week, for work completed the week prior.