Nibedita Bisoyi

AI Trainer and Data Annotation Specialist with 4+ years of experience in large-scale multimodal AI projects including image tagging, LLM evaluation, prompt engineering, and multilingual transcription. Proven expertise in improving AI model accuracy, safety, and linguistic quality through structured feedback, annotation, and QA. Proficient in English, Hindi, and Odia, with a strong commitment to ethical AI and high-quality dataset development.

AI Data Trainer at Invisible Technologies (2025-08 – Present)

• Evaluate and score AI model outputs for accuracy, safety, reasoning, and bias.
• Perform prompt-response validation and structured feedback for LLM training.
• Detect hallucinations, policy violations, and logical errors in AI responses.
• Generate and refine prompts to improve model alignment and output quality.
• Conduct multilingual evaluation in English, Hindi, and Odia.
• Contribute to reinforcement learning and supervised fine-tuning pipelines

Image Data Annotation & Tagging Specialist at iMerit (2025-11 – Present)

• Perform image tagging, object labeling, and classification for computer vision models.
• Annotate large-scale visual datasets for detection, recognition, and contextual understanding.
• Maintain high accuracy and consistency across production workflows.
• Review and validate annotation quality for training-grade datasets.
• Support enterprise-level AI and machine learning systems

Business Data Analyst – AI & Data Annotation at Turing (2025-07 – Present)

• Analyze and structure datasets for machine learning and generative AI workflows.
• Perform prompt evaluation, AI content writing, and linguistic QA.
• Execute data annotation, transcription, classification, and review tasks.
• Identify output patterns to improve model behavior and relevance.
• Collaborate with engineering teams to improve dataset quality and efficiency.

Language AI Expert & Data Annotator at Appen (2023-01 – Present)

• Annotate image, video, and text datasets for AI training.
• Review AI-generated responses for fluency, correctness, and cultural relevance.
• Support NLP and LLM fine-tuning through supervised evaluation.
• Perform Instagram tagging and social-media content classification.

AI Trainer & Prompt Specialist at Outlier (2023-01 – 2025-12)

• Designed and generated high-quality prompts for AI models across Mathematics, Science, Law, and Humanities.
• Evaluated and reviewed AI-generated responses for logical accuracy, factual correctness, and instruction-following.
• Performed chain-of-thought validation, error detection, and quality scoring for large language models.
• Labeled, ranked, and refined model outputs to support supervised fine-tuning and reinforcement learning.
• Identified hallucinations, bias, and safety risks in AI responses and provided corrective feedback.
• Contributed to improving reasoning depth, language quality, and domain accuracy of production AI systems.

MBA in International Business & HRM – BPUT, Odisha (2021-01 – 2024-12)

BA in English – Berhampur University (2015-01 – 2018-12)