Haryanvi Dialect Specialist

About the role

This position channels your deep Haryanvi linguistic expertise directly into the evaluation and refinement of language models designed for Haryanvi speakers. You will define the specific analysis scope required to assess how these models handle the nuances of the language. The core responsibility involves converting this expertise into tangible model evaluation data that highlights strengths and failure points. You will own the documentation of error traces, particularly in verb forms, cases, word order, and stylistic choices encountered during testing. This role systematically assesses model outputs to generate concrete suggestions for improving prompts and evaluation metrics. Success in this position depends on your ability to provide clear, metacognitive explanations of your linguistic analysis. The work directly shapes the quality and accuracy of data used in AI training for Haryanvi language applications.

Key facts

What you'll do

Design and execute conversational tests with language models focused on Haryanvi scenarios to uncover weaknesses in verb conjugation and morphology.

Document detailed error traces from these model interactions, emphasizing issues in case usage, word order, and stylistic appropriateness.

Analyze the documented failures to generate specific, actionable suggestions for refining model prompts and evaluation frameworks.

Systematically assess model outputs against linguistic standards to propose improvements in both prompt design and metric definitions.

Collaborate closely with data scientists and engineers to translate linguistic insights into concrete changes in training data evaluation processes.

Apply deep mastery of Haryanvi grammar, syntax, morphology, phonology, semantics, and pragmatics to judge the correctness and fluency of model generations.

Utilize clear, metacognitive reasoning to explain every finding, ensuring that your analysis is reproducible and well-understood by technical teams.

Employ a secure computer and high-speed internet to perform all required evaluations and communications reliably.

Interpret experiment results and provide concrete feedback that directly influences the development of better language models for Haryanvi speakers.

Build a portfolio of past analyses and evaluations that demonstrates your practical ability to assess language models, often valued more than formal degrees.

Engage with complex linguistic scenarios to identify subtle errors that could impact user experience for Haryanvi speakers.

Support the creation of robust evaluation metrics by providing real-world linguistic data from model interaction testing.

Contribute to the overall improvement of AI training data by ensuring linguistic accuracy and cultural relevance through meticulous evaluation.

Participate in typical data interview processes, which may include SQL or coding exercises, statistics questions, and case studies related to language evaluation.

Requirements

The posting states a pay range of $6 to $65.

You must possess a Bachelor's degree in Haryanvi language, linguistics, or a closely related field, though equivalent practical experience can substitute if formal study in this specific area is absent.

Mastery of Haryanvi grammar, syntax, morphology, phonology, semantics, and pragmatics must remain absolute and non-negotiable for success in this role.

You must be capable of providing clear, metacognitive explanations that reveal your reasoning process for documenting linguistic analysis.

A secure computer and high-speed internet connectivity are essential prerequisites for performing the core job functions.

You should be prepared to engage in data interview steps that commonly include technical exercises, case studies, and discussions about past projects.

Candidates must demonstrate an understanding of how language structure impacts model testing within AI training data pipelines.

Success in this role hinges on your ability to perform reproducible tracing of errors and communicate findings effectively to cross-functional teams.

Practical notes

A secure computer and high-speed internet are necessary for performing the work.

Typical interview steps

Data interviews commonly include a SQL or coding exercise, a statistics question, and a case study. Candidates may be asked to design a metric, interpret an experiment, or build a small model. Some companies give a take-home analysis. Expect questions about past projects and the business impact of your work. Interviewers often evaluate how you communicate uncertainty and business impact, not only the math. Bringing a clean write-up of a past analysis to the interview is well received.