Ankur Kumar

Key Skills

Software

Mindrift

Top Subject Matter

AI-powered automation workflows

deep learning

AI model evaluation

Top Data Types

Image

Text

Video

Top Task Types

Classification

Freelancer Overview

AI and ML Internship (AI Data Labeling/Annotation). Brings 2+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal, Proprietary Tooling, and Mindrift. Education includes Bachelor of Technology, Technocrats Institute of Technology (2024). AI-training focus includes data types such as Image, Text, and Video and labeling workflows including Classification, Evaluation, and Rating.

Labeling Experience

AI and ML Internship (AI Data Labeling/Annotation)

ImageClassification

During my AI and ML internship at CDAC, I participated in the enhancement of annotation tools for machine learning projects. I focused on model enhancement, dataset preparation, and the evaluation of deep learning applications. My work directly contributed to the improvement of annotation accuracy and efficiency in data labeling workflows. • Collaborated on dataset creation and annotation for deep learning models. • Used and improved internal open-source annotation tools compatible with multiple ML frameworks. • Enhanced annotation accuracy by 20% and accelerated training cycles by 25%. • Supported workflows that serviced over 100 users in real-world AI deployments.

During my AI and ML internship at CDAC, I participated in the enhancement of annotation tools for machine learning projects. I focused on model enhancement, dataset preparation, and the evaluation of deep learning applications. My work directly contributed to the improvement of annotation accuracy and efficiency in data labeling workflows. • Collaborated on dataset creation and annotation for deep learning models. • Used and improved internal open-source annotation tools compatible with multiple ML frameworks. • Enhanced annotation accuracy by 20% and accelerated training cycles by 25%. • Supported workflows that serviced over 100 users in real-world AI deployments.

2025 - Present

Freelance AI Tutor (Prompt Evaluation/Data Labeling)

MindriftText

As a freelance AI tutor for Mindrift, I worked on improving AI model responses through quality checks and reasoning evaluation. My tasks included prompt refinement and assessment of model outputs to enhance alignment and helpfulness. This short-term engagement focused on the evaluation and improvement of AI-generated text datasets. • Conducted prompt refinement for large language model outputs. • Evaluated and rated AI-generated responses for reasoning quality. • Performed quality checks to support model improvement cycles. • Used Mindrift as the primary platform for all evaluation activities.

As a freelance AI tutor for Mindrift, I worked on improving AI model responses through quality checks and reasoning evaluation. My tasks included prompt refinement and assessment of model outputs to enhance alignment and helpfulness. This short-term engagement focused on the evaluation and improvement of AI-generated text datasets. • Conducted prompt refinement for large language model outputs. • Evaluated and rated AI-generated responses for reasoning quality. • Performed quality checks to support model improvement cycles. • Used Mindrift as the primary platform for all evaluation activities.

2025 - 2025

Project Lead (AI Annotation Platform)

ImageClassification

I developed a full-stack AI annotation platform, AnnoFlow, facilitating both image and video labeling for machine learning workflows. The platform supported multi-user real-time collaboration and significantly reduced annotation timelines. This project improved the efficiency and scalability of data labeling for computer vision applications. • Designed and implemented data labeling features for images and videos. • Enabled real-time, collaborative annotation to boost productivity. • Integrated the platform with ML pipelines to accelerate training. • Contributed to the deployment and usability of an internal annotation tool.

I developed a full-stack AI annotation platform, AnnoFlow, facilitating both image and video labeling for machine learning workflows. The platform supported multi-user real-time collaboration and significantly reduced annotation timelines. This project improved the efficiency and scalability of data labeling for computer vision applications. • Designed and implemented data labeling features for images and videos. • Enabled real-time, collaborative annotation to boost productivity. • Integrated the platform with ML pipelines to accelerate training. • Contributed to the deployment and usability of an internal annotation tool.

Not specified

Education

Work History