Vineet S.

Key Skills

Software

Other

Data Annotation Tech

AWS SageMaker

OpenCV AI Kit (OAK)

Top Subject Matter

Computer vision (face recognition, object detection)

Real-time video/screen data collection for AI development support

Top Data Types

Image

Text

Video

Top Task Types

Object Detection

Data Collection

Classification

Freelancer Overview

Personal AI Assistant (2019-2021) and Object Detection System (2019-2021). Brings 7+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include OpenCV and Other. Education includes Bachelor of Technology, Chandigarh University (2029) and Bachelor of Science, IIT Guwahati (2029). AI-training focus includes data types such as Image and labeling workflows including Object Detection and Data Collection.

Labeling Experience

Web Meeting Platform (Jitsi Meet Based) (2020-2021)

OtherImageData Collection

Designed and deployed an AI-adjacent vision system project using a secure WebRTC-based web meeting platform stack for low-latency streaming. While not explicitly an annotation role, the work involved configuring room creation, participant management, and real-time media handling that supports data generation for training and evaluation pipelines. The platform architecture can be leveraged to collect synchronized multi-user visual data for subsequent AI model development. • Configured WebRTC/Jitsi components for real-time screen sharing and chat. • Optimized performance for low-latency streaming suitable for dataset capture. • Implemented participant and room management features for controlled data collection. • Deployed on Google Cloud and AWS to support repeatable runs for data capture.

2020 - 2021

Personal AI Assistant (2019-2021) and Object Detection System (2019-2021)

ImageObject Detection

Developed a personal AI assistant and computer vision pipelines using OpenCV and TensorFlow, focusing on real-time face recognition and vision-based triggers. Implemented object detection using pre-trained CNN models over webcam video with bounding-box visualization to support labeled outputs for detected objects. Built interactive command handling connected to automation tasks to validate model behavior in practical scenarios. • Used webcam feed as the input stream for real-time visual inference. • Applied CNN-based detection with bounding boxes to visualize detected object locations. • Designed trigger logic based on vision results and basic automation workflows. • Included desktop automation and voice/text command execution for end-to-end testing.

2019 - 2021

Education

Work History