AI Data Specialist - Turkish

About the role

You will own the meticulous evaluation and refinement of AI-generated Turkish content to ensure it meets rigorous quality and safety standards. This role places you at the intersection of technology and linguistics, where your native-level judgment will directly influence the behavior of AI models. You will be responsible for conducting detailed comparative analyses to determine the superiority of responses, ensuring that outputs are accurate and contextually appropriate. The position requires a high degree of ownership over data integrity, where you will identify and tag specific objects within multimedia and text-based materials. You will engage in systematic counting and classification tasks to quantify data quality and model performance. This is a critical function that supports the development of safer, more reliable AI systems for Turkish-speaking users. By participating in this project, you will contribute to shaping the future landscape of artificial intelligence and its responsible deployment. Your daily work will involve direct interaction with diverse content types, requiring a keen eye for detail and consistency.

Key facts

What you'll do

Conduct exhaustive evaluations of AI-generated Turkish text to identify factual inaccuracies, grammatical errors, and logical inconsistencies.

Perform intricate pairwise comparisons between multiple AI responses to determine which is more accurate, coherent, and contextually relevant.

Execute systematic counting tasks to quantify specific linguistic patterns, errors, or data occurrences within provided datasets.

Engage in comprehensive data collection by sourcing and gathering relevant Turkish language content from various digital environments.

Apply specialized object tagging and labeling techniques to audio, video, image files, and structured data to prepare them for machine learning pipelines.

Execute rigorous data evaluation and quality assurance protocols to ensure that all annotated materials meet strict accuracy standards.

Analyze complex data sets to preprocess information, making it suitable for training, validation, and testing phases of model development.

Assign precise labels and metadata to content across diverse formats, ensuring that categorization supports effective machine learning workflows.

Monitor and document the performance of AI models through continuous assessment of their output against predefined quality benchmarks.

Collaborate with cross-functional teams to align data annotation strategies with evolving project requirements and AI safety goals.

Implement meticulous review processes to verify the correctness and reliability of labeled data before it is used in model training.

Adapt to shifting project priorities by taking on varied data-related responsibilities as the needs of the AI development cycle dictate.

Utilize advanced English language skills to interpret source materials and guidelines, ensuring alignment with international quality standards.

Maintain a high level of organization and time management to meet deadlines and manage large volumes of data efficiently.

Requirements

Must be a native speaker of Turkish with flawless comprehension and expression in both written and spoken forms.

Must possess fluent or advanced English proficiency, demonstrating a certified level of B2, C1, or C2 on standardized language assessment scales.

Must reside in or be physically located in Bulgaria, specifically in the Sofia region, to comply with operational requirements.

Must have legal authorization to work on a freelance or contract basis within Bulgaria, including valid tax registration if required.

Must be available to commit a minimum of 10 hours per week to ensure continuity and momentum in data evaluation tasks.

Must be capable of working independently with a high degree of self-motivation and discipline in a fully remote environment.

Must possess strong analytical skills to discern subtle differences in meaning, tone, and accuracy between data samples.

Must demonstrate a proven track record of reliability and attention to detail, with the ability to consistently meet strict quality benchmarks.

Nice to have

Prior experience in machine learning tasks, including data preprocessing, model evaluation, or participation in AI training cycles.

Previous work in data collection, data evaluation, and data annotation and labeling roles within the technology sector.

Familiarity with standard AI & data capabilities frameworks and an understanding of how training data impacts model behavior.

Experience working with audio, video, images, and other non-textual content formats for data labeling purposes.

Practical notes

The engagement is part-time, requiring a commitment of 10+ hours per week.

The schedule is flexible, allowing the contractor to work whenever they want within their committed hours.

The rate is fixed at 9 USD per hour, reflecting the specific cost of living and market rates for the location.

This opportunity is ideally suited for students, recent graduates, stay-at-home parents, and gig workers seeking supplemental income.

The position offers timely payments, ensuring a reliable and predictable cash flow for contractors.

Work is to be performed remotely from the contractor's home environment, eliminating the need for daily commuting.

There is no specified end date listed for this role, indicating an ongoing need for support.