Hire this AI Trainer
Sign in or create an account to invite AI Trainers to your job.
Image
Video
Audio
Object Detection
Bounding Box
TranscriptionI am an AI data annotation specialist with 1.5 years of hands-on experience supporting multimodal AI training project for Garmr, a U.S.-based startup, involving image and video data. I worked with proprietary annotation tools and MakeSense.ai to perform object detection, bounding box annotation, visual classification, and structured label assignment based on detailed annotation guidelines. My work also included dataset curation, quality-focused review, and handling ambiguous or edge cases to maintain annotation consistency and dataset reliability. In addition to visual annotation, I have experience contributing to multilingual AI training workflows through transcription, audio data collection, linguistic validation, proofreading, and structured data preparation in Belarusian, Russian, and English. My background also includes over 10 years of museum collection database management focused on archaeological artifacts, where I developed strong skills in systematic classification, precise visual analysis, and maintaining accuracy across large structured datasets.
Audio
Data Collection- Recorded audio data in Belarusian and English for machine learning datasets. - Translated texts from Russian into English for project support. - Performed editing, proofreading, and fact-checking in Belarusian, ensuring linguistic accuracy and consistency.
VideoObject Detection- Annotated and classified image and video data for multimodal AI training using proprietary annotation software and MakeSense.ai. - Performed object detection, bounding box annotation, visual classification, and structured label assignment based on detailed project guidelines. - Ensured annotation consistency and dataset reliability through quality-focused review and careful handling of ambiguous or edge cases. - Curated and organized video datasets for annotation workflows according to project requirements. - Identified issues in the internal annotation platform, documented bugs for developers, and verified fixes after implementation to ensure stable workflow functionality.
AudioTranscription- Transcribed Belarusian audio files into text using Loft 2.0 with high accuracy and attention to linguistic detail. - Edited and revised transcriptions to ensure compliance with project-specific quality and formatting standards. - Performed fact-checking and source verification to validate names, terminology, and contextual information. - Followed detailed transcription guidelines while maintaining consistency and meeting project deadlines.
PhD Candidate, Archaeology
Master’s Degree, History (Archaeology)
Belarusian Transcriber (Freelance/Contract)
Head of the Department of Archaeology, Numismatics, and Weaponry / Archaeology Collection Curator / Researcher
