Vishnupriya

Data Annotation Specialist with 2+ years of experience in AI/ML data labeling across NLP, ASR, computer vision, and LiDAR domains. Skilled in text, audio, image, video, and 2D/3D point cloud annotation, QC, and multilingual dataset creation (Tamil, English). Experienced in LLM support through translation and localization of English content into Tamil.

Proficient in Label Studio, Dataloop, Amazon SageMaker, and data validation workflows. Experienced in dataset curation and annotation QA for large-scale multimodal AI systems.

Data Annotator - Cyllid Technologies Private Limited (Skit.Ai) - Bangalore, India

(2026-01)

• Data Generation & Recording: Generated and recorded Tamil user utterances to train conversational AI voice bots.
• Intent Tagging: Performed intent tagging using Label Studio to improve NLP model accuracy.
• NLP Bot Testing: Evaluated AI responses for accuracy, tone, and conversational flow.
• Live Call Monitoring: Monitored live calls for ASR accuracy, latency, and interaction issues.
• Post-Live Call Monitoring (CRR): Conducted post-call audits (CRR) to identify errors, drop-offs, and bugs.
• Prompt Translation: Translated and localized chatbot prompts from English to Tamil for multilingual support.

Process Associate - Kriya NextWealth Private Limited - Bangalore, India

(2025-09 - 2026-01)

• Image Annotation: Annotated and validated large-scale product images to ensure correct alignment with titles and descriptions for computer vision and e-commerce AI models.
• Video Annotation: Performed long-form video content analysis and scene labeling to support object detection, product categorization, and machine learning training datasets.
• Image Description Translation & LLM Support: Translated and localized image metadata and English content into Tamil to support multilingual AI datasets and improve LLM-based model understanding and regional language accuracy.

Audio Annotator - OLA - Krutrim SI Designs - Bangalore, India

(2025-03 - 2025-08)

• Speech-to-Text & Audio Annotation: Transcribed Tamil and multilingual audio recordings with 98%+ accuracy and performed audio annotation for emotion, tone, speaker role, and intent to support ASR, NLP, speech recognition, and sentiment classification models.
• Audio Validation & Transcript Alignment: Validated YouTube audio recordings for accurate alignment with transcripts to support TTS and ASR model fine-tuning.

Data Annotator - Objectways Technologies - Coimbatore, India

(2023-05 - 2024-02)

• Text Annotation: Selected the best answer from question-answer pairs based on relevance and accuracy for NLP model training
• Audio Annotation: Curated audio samples based on given topics for speech recognition dataset development
• Video Annotation (BDX): Validated ad videos against transcripts, including speech and visual elements (logos, company names) for multimodal AI models
• Blueprint Data Extraction: Extracted key information from multi-page documents and created structured summaries highlighting important business data
• LiDAR Annotation: Performed 2D and 3D point cloud annotation for object detection and computer vision model training

Bachelor of Science - Computer Science - LRG Government Arts College for Women (2019-06 - 2022-06)