Vivekanand Chimman

Key Skills

Software

AWS SageMaker

Appen

Label Studio

Roboflow

Remotasks

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Image

Text

Video

Top Task Types

Bounding Box

Classification

Computer Programming Coding

Data Collection

Freelancer Overview

Proficient in data labeling and annotation for object detection, action recognition, phrase grounding, and video clip extraction, leveraging tools like Label Studio, SAM, and Florence-2 to create high-quality structured datasets and accelerate the annotation process. Experienced in bounding box, polygon, and segmentation-based labeling to prepare robust datasets for training and fine-tuning deep learning models. In addition, I have freelance experience with platforms like Appen (CrowdGen), where I worked on audio transcription, and Remotasks, where I handled code and prompt reviews and model response reviews.

Labeling Experience

Text response check

AppenTextEvaluation Rating

Worked on an Appen project to evaluate whether the given audio clips were pronounced correctly.

Worked on an Appen project to evaluate whether the given audio clips were pronounced correctly.

2025

Transcription

AppenAudioText Generation

Worked on an Appen project to transcribe the given audio clips accurately.

Worked on an Appen project to transcribe the given audio clips accurately.

2025 - 2025

Image Annotation

RoboflowImageBounding Box

In this project, I performed data labeling and annotation for training YOLO-based object detection models used in a prompt-driven video clipping tool. The task involved annotating thousands of video frames extracted from long-duration videos, focusing on object detection, action instances, and region-specific bounding boxes. Using tools like Roboflow, Label Studio and others, I ensured high-quality annotations by adhering to strict IoU accuracy thresholds, class consistency checks, and cross-verification processes. The labeled datasets were later used to train custom YOLO models, which powered the automated detection of objects and actions within videos. These trained models were integrated into the video clipping pipeline, enabling the system to identify and extract relevant video segments based on user prompts with high precision.

In this project, I performed data labeling and annotation for training YOLO-based object detection models used in a prompt-driven video clipping tool. The task involved annotating thousands of video frames extracted from long-duration videos, focusing on object detection, action instances, and region-specific bounding boxes. Using tools like Roboflow, Label Studio and others, I ensured high-quality annotations by adhering to strict IoU accuracy thresholds, class consistency checks, and cross-verification processes. The labeled datasets were later used to train custom YOLO models, which powered the automated detection of objects and actions within videos. These trained models were integrated into the video clipping pipeline, enabling the system to identify and extract relevant video segments based on user prompts with high precision.

2025 - 2025

Programming

Scale AIComputer Code ProgrammingComputer Programming Coding

Reviewed AI generated Python code

Reviewed AI generated Python code

2023 - 2023

Education

Work History