Hire this AI Trainer
Sign in or create an account to invite AI Trainers to your job.
Mindrift
Label Studio
Toloka
Text
Image
Video
AudioAI Trainer (Mindrift). Brings 2+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Mindrift and Label Studio. AI-training focus includes data types such as Video, Audio, Image, Geospatial and Text and labeling workflows including Data Collection, Segmentation, and Entity (NER) Classification.
Performed land cover mapping and building footprints on high-resolution satellite and drone images.
Captured and labeled egocentric video data following rigorous annotation protocols to support AI dataset development. Participated in data collection for structured audio and video inputs, adhering to accuracy and privacy standards.
Captured and labeled egocentric video data following rigorous annotation protocols to support AI dataset development. Participated in data collection for structured audio and video inputs, adhering to accuracy and privacy standards. Evaluated shopping application AI agent outputs for consistency and usability in real-world scenarios. • Executed data annotation for video and audio samples • Ensured high-fidelity egocentric data quality • Conducted app agent evaluation for reliability metrics • Contributed to building robust multimodal training sets
As an AI Data Trainer, I specialized in Reinforcement Learning from Human Feedback (RLHF), model alignment, and quality assurance. My key responsibilities included optimizing language model performance, engineering adversarial prompts, and evaluating outputs for reasoning and factual accuracy. I contributed to ground-truth benchmarking by crafting ideal responses and curating multi-modal datasets for model training. • Designed and applied MECE-based rubrics for consistent model evaluation. • Curated and annotated complex image-text datasets for visual recognition tasks. • Created chain-of-thought benchmarks and gold-standard responses for fine-tuning. • Executed red teaming and root-cause analysis to identify and mitigate AI hallucinations.
Managed high-volume end-to-end annotation tasks on images, focusing on retail products, vehicles, and facial landmarks for computer vision development. Applied semantic and instance segmentation techniques to distinguish overlapping and similar objects clearly. Used Label Studio for systematic project tracking and ensured pixel-level precision in labeled data. • Improved object recognition accuracy for model training • Handled large-scale annotation with consistent workflows • Performed peer-reviewed QC for annotation reliability • Enhanced computer vision datasets for diverse tasks
Bachelor of Medical Laboratory Science, Medical Laboratory Sciences
AI Data Contributor
AI Trainer
