For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
Deepak Giri

Deepak Giri

AI Research Annotator - AI Training Workflows

INDIA flag
South West Delhi, India
$20.00/hrEntry LevelOther

Key Skills

Software

Other

Top Subject Matter

No subject matter listed

Top Data Types

VideoVideo
ImageImage
AudioAudio

Top Label Types

Question Answering
Emotion Recognition
Object Detection
Bounding Box
Action Recognition
Tracking
Evaluation Rating
Transcription

Freelancer Overview

I am an experienced data annotator and AI training data specialist with a strong background in mathematics and analytical reasoning. My work has involved prompt annotation, LLM response evaluation, and image review, where I ensured accuracy, relevance, and adherence to detailed guidelines. I have contributed to large-scale AI training workflows by performing content evaluation, video-based annotation, and question generation to improve model understanding. My experience includes identifying logical errors, bias, and inconsistencies across text and image datasets, always maintaining high quality standards. I am comfortable with remote, guideline-driven tasks and excel at delivering precise, consistent results to support AI and machine learning projects.

Entry LevelEnglish

Labeling Experience

Data Annotator

OtherAudioTranscription
Performed audio transcription and quality validation to convert spoken content into accurate, structured text for AI training datasets. Reviewed transcripts against original audio recordings to identify and correct omissions, misinterpretations, and formatting inconsistencies. Ensured clarity, contextual accuracy, and adherence to predefined transcription guidelines. Maintained high precision standards through systematic review and error correction to enhance dataset reliability for speech recognition and language modeling systems.

Performed audio transcription and quality validation to convert spoken content into accurate, structured text for AI training datasets. Reviewed transcripts against original audio recordings to identify and correct omissions, misinterpretations, and formatting inconsistencies. Ensured clarity, contextual accuracy, and adherence to predefined transcription guidelines. Maintained high precision standards through systematic review and error correction to enhance dataset reliability for speech recognition and language modeling systems.

2025

Data Annotator

OtherVideoEmotion RecognitionObject Detection
Performed video based data annotation using object picker and bounding box tools to support computer vision and multimodal AI training. Identified and tagged objects within video frames while capturing precise spatial coordinates. Applied color picker tools to annotate visual attributes and object level features. Conducted action recognition by labeling human activities and motion patterns across frames, and performed tracking when required to maintain object continuity over temporal sequences. Contributed structured annotations aligned with predefined taxonomies and ensured high accuracy through strict adherence to quality guidelines and validation protocols.

Performed video based data annotation using object picker and bounding box tools to support computer vision and multimodal AI training. Identified and tagged objects within video frames while capturing precise spatial coordinates. Applied color picker tools to annotate visual attributes and object level features. Conducted action recognition by labeling human activities and motion patterns across frames, and performed tracking when required to maintain object continuity over temporal sequences. Contributed structured annotations aligned with predefined taxonomies and ensured high accuracy through strict adherence to quality guidelines and validation protocols.

2025

Data Annotator

OtherImageBounding BoxEmotion Recognition
Performed image annotation using object picker and bounding box tools to support computer vision model training. Identified and tagged objects while capturing precise X and Y coordinate values for accurate spatial localization. Applied bounding box techniques to detect human subjects and environmental elements. Conducted emotion recognition by labeling facial expressions and performed action recognition by identifying visible human activities within static frames. Generated structured descriptions linking object names with coordinate data to enhance supervised learning datasets. Ensured strict adherence to annotation guidelines and quality standards.

Performed image annotation using object picker and bounding box tools to support computer vision model training. Identified and tagged objects while capturing precise X and Y coordinate values for accurate spatial localization. Applied bounding box techniques to detect human subjects and environmental elements. Conducted emotion recognition by labeling facial expressions and performed action recognition by identifying visible human activities within static frames. Generated structured descriptions linking object names with coordinate data to enhance supervised learning datasets. Ensured strict adherence to annotation guidelines and quality standards.

2025

Video Question Answering Evaluator

OtherVideoQuestion AnsweringEmotion Recognition
Analyzed short video clips to design challenging, context grounded questions aimed at testing multimodal AI comprehension. Generated high difficulty queries based on visual details, temporal sequences, and scene level context. Submitted videos and formulated questions to Gemini Pro to evaluate model reasoning, grounding accuracy, and response completeness. Identified failure cases where the model produced incorrect, incomplete, or hallucinated answers. Validated correct responses through manual review and documented verified ground truth answers in structured spreadsheets for quality tracking.

Analyzed short video clips to design challenging, context grounded questions aimed at testing multimodal AI comprehension. Generated high difficulty queries based on visual details, temporal sequences, and scene level context. Submitted videos and formulated questions to Gemini Pro to evaluate model reasoning, grounding accuracy, and response completeness. Identified failure cases where the model produced incorrect, incomplete, or hallucinated answers. Validated correct responses through manual review and documented verified ground truth answers in structured spreadsheets for quality tracking.

2025 - 2025

Education

D

DM College of Science Manipur University

Bachelor of Science, Mathematics (Hons)

Bachelor of Science
2012 - 2016

Work History

C

Chegg India Pvt Ltd

Subject Matter Expert

Remote
2020 - Present