Alen Kurian

Key Skills

Software

Labelbox

Data Annotation Tech

Top Subject Matter

No subject matter listed

Top Data Types

Text

Audio

Top Label Types

Classification

Segmentation

Question Answering

Freelancer Overview

I am an AI Specialist and Data Annotation Expert with over five years of hands-on experience in AI training data development, evaluation, and optimization. I have contributed to large-scale projects involving large language models (LLMs), natural language processing (NLP), computer vision, robotics, and generative AI systems. My expertise includes text classification, named entity recognition (NER), sentiment analysis, image and video annotation, multimodal data preparation, prompt engineering, and structured model evaluation. I am highly skilled in Python, annotation platforms such as CVAT and LabelImg, and machine learning frameworks including TensorFlow and PyTorch. My work focuses on ensuring data precision, reasoning validation, safety alignment, and gold-standard dataset creation to enhance AI model accuracy and reliability. With extensive experience collaborating remotely with global AI teams, I consistently deliver high-quality, scalable training data solutions for advanced machine learning applications.

Labeling Experience

Large Language Model Evaluation Framework

LabelboxTextQuestion Answering

Developed structured evaluation metrics for grading reasoning, coherence, factuality, and safety of AI-generated responses.

Developed structured evaluation metrics for grading reasoning, coherence, factuality, and safety of AI-generated responses.

2025

Music Genre Classification Dataset Labeling

Data Annotation TechAudioSegmentationClassification

Annotated music clips from the GTZAN dataset by genre, creating labeled spectrograms for CNN model training. Performed audio segmentation to isolate key patterns, ensuring balanced representation across 10 genres. Quality assurance involved multiple verification rounds to eliminate mislabels.

Annotated music clips from the GTZAN dataset by genre, creating labeled spectrograms for CNN model training. Performed audio segmentation to isolate key patterns, ensuring balanced representation across 10 genres. Quality assurance involved multiple verification rounds to eliminate mislabels.

2024 - 2024

Credit Card Fraud Detection Dataset Annotation

LabelboxTextClassification

Annotated and validated large-scale credit card transaction datasets for supervised machine learning models, labeling entries as "Fraud" or "Non-Fraud." Ensured data balance between classes, handled missing values, and maintained strict quality control through multiple review passes. The labeled data was used to train Random Forest and Logistic Regression models, achieving high precision and recall in fraud detection.

Annotated and validated large-scale credit card transaction datasets for supervised machine learning models, labeling entries as "Fraud" or "Non-Fraud." Ensured data balance between classes, handled missing values, and maintained strict quality control through multiple review passes. The labeled data was used to train Random Forest and Logistic Regression models, achieving high precision and recall in fraud detection.

2024 - 2024

Education

Work History