Lakshay Bhardwaj

Key Skills

Software

Appen

Clickworker

Scale AI

Data Annotation Tech

Top Subject Matter

No subject matter listed

Top Data Types

Text

Image

Audio

Top Task Types

Classification

Evaluation Rating

Data Collection

Bounding Box

Object Detection

Audio Recording

Freelancer Overview

I am an experienced AI Data Annotator and Quality Auditor with a strong background in labeling and auditing data for Vision-Language Models (VLMs) and Large Language Models (LLMs). My work spans multimodal annotation—including text, image, video, and search relevance tasks—as well as detailed evaluation of reasoning accuracy, semantic correctness, and hallucination risk in AI outputs. I have contributed to high-priority projects for major platforms, ensuring dataset quality through strict guideline enforcement, audit workflows, and error analysis. My skills include image annotation, bounding boxes, entity recognition, style consistency checks, and meta-level quality assurance. I am adept at both hands-on annotation and quality management, bringing a rigorous and detail-oriented approach to every stage of the AI training data pipeline.

Labeling Experience

Audio Quality Evaluation for AI Training

Data Annotation TechAudioEvaluation RatingAudio Recording

Evaluated and labeled 50,000 audio clips to improve speech-to-text performance in customer support AI. Focused on identifying transcription errors and ensuring adherence to linguistic rules, resulting in a 15% improvement in model accuracy.

Evaluated and labeled 50,000 audio clips to improve speech-to-text performance in customer support AI. Focused on identifying transcription errors and ensuring adherence to linguistic rules, resulting in a 15% improvement in model accuracy.

2023 - 2024

Text Annotation for Sentiment Analysis

AppenTextClassificationEvaluation Rating

Conducted text classification tasks to label sentiment for 10,000+ social media comments. Ensured precise sentiment categorization (positive, negative, neutral) to train NLP models. Achieved 98% accuracy in evaluations by adhering to quality assurance guidelines.

Conducted text classification tasks to label sentiment for 10,000+ social media comments. Ensured precise sentiment categorization (positive, negative, neutral) to train NLP models. Achieved 98% accuracy in evaluations by adhering to quality assurance guidelines.

2022 - 2022

Image Annotation for Autonomous Driving Models

Scale AIImageBounding BoxObject Detection

Labeled over 10,000 images for vehicle and pedestrian detection by creating 3D bounding boxes. Supported autonomous driving models by identifying road signs, lane markings, and dynamic objects. Ensured 95% accuracy through collaborative quality checks.

Labeled over 10,000 images for vehicle and pedestrian detection by creating 3D bounding boxes. Supported autonomous driving models by identifying road signs, lane markings, and dynamic objects. Ensured 95% accuracy through collaborative quality checks.

2021 - 2022

Data Collection for Natural Language Processing (NLP)

ClickworkerTextClassificationData Collection

Collected and labeled user search queries for training an e-commerce recommendation system. Categorized 30,000+ search terms into relevant product categories, achieving 95% relevancy accuracy in client evaluations.

Collected and labeled user search queries for training an e-commerce recommendation system. Categorized 30,000+ search terms into relevant product categories, achieving 95% relevancy accuracy in client evaluations.

2019 - 2019

Education

Work History