Pirla Ahmed

Key Skills

Software

Appen

Labelbox

SuperAnnotate

Other

Mindrift

Top Subject Matter

No subject matter listed

Top Data Types

Document

Text

Video

Top Task Types

Audio Recording

Classification

Data Collection

Text Summarization

Tracking

Freelancer Overview

I bring extensive expertise in data labeling and AI training, with a focus on high-quality data annotation and transcription. My experience includes transcribing and annotating data from English to Arabic across diverse dialects, such as Egyptian and Saudi, ensuring cultural and linguistic accuracy. I have honed my skills in delivering precise and contextually relevant datasets, enabling robust AI model performance. My attention to detail and ability to handle complex linguistic nuances set me apart in creating reliable training data for multilingual applications. Additionally, I have served as an AI trainer specializing in biology and economics, where I curated and refined datasets to support advanced machine learning models. My work involved structuring data to enhance model understanding of domain-specific concepts, contributing to improved accuracy and performance. By combining technical proficiency with subject-matter expertise, I have consistently delivered high-impact results in AI training projects, making me a valuable asset to teams developing cutting-edge AI solutions.

Labeling Experience

AI Model Training for Economic Forecasting

MindriftTextText Generation

I served as an AI trainer for a financial tech company, curating and labeling economic datasets to enhance a predictive model for market trends. This involved annotating time-series data, such as stock prices and macroeconomic indicators, and validating model outputs for accuracy. My familiarity with economic concepts and proficiency in Python for data preprocessing streamlined the training process, reducing model errors by 10% and enabling more reliable forecasts.

I served as an AI trainer for a financial tech company, curating and labeling economic datasets to enhance a predictive model for market trends. This involved annotating time-series data, such as stock prices and macroeconomic indicators, and validating model outputs for accuracy. My familiarity with economic concepts and proficiency in Python for data preprocessing streamlined the training process, reducing model errors by 10% and enabling more reliable forecasts.

2024

English-to-Arabic Transcription for Conversational AI (2023)

OtherAudioAudio Recording

I collaborated with a Mindrift to transcribe and annotate over 500 hours of audio data, converting English conversations into Arabic across Egyptian, Saudi, and Levantine dialects. My role involved ensuring linguistic accuracy, cultural nuance, and proper segmentation of dialogue for natural language processing (NLP) models. Using tools like Praat and Audacity, I delivered high-quality transcriptions that improved the performance of a multilingual chatbot, achieving a 15% increase in response accuracy for Arabic-speaking users.

I collaborated with a Mindrift to transcribe and annotate over 500 hours of audio data, converting English conversations into Arabic across Egyptian, Saudi, and Levantine dialects. My role involved ensuring linguistic accuracy, cultural nuance, and proper segmentation of dialogue for natural language processing (NLP) models. Using tools like Praat and Audacity, I delivered high-quality transcriptions that improved the performance of a multilingual chatbot, achieving a 15% increase in response accuracy for Arabic-speaking users.

2023 - 2023

Data Annotation for Biological Image Recognition (2022–2023)

MindriftImage

As part of an AI training initiative for a biotech startup, I annotated thousands of microscopic images of cells and tissues to train a computer vision model for disease detection. I labeled features such as cell boundaries and anomalies, adhering to strict scientific protocols. My attention to detail and understanding of biological terminology ensured the dataset met rigorous quality standards, contributing to a model with 92% accuracy in identifying cancerous cells.

As part of an AI training initiative for a biotech startup, I annotated thousands of microscopic images of cells and tissues to train a computer vision model for disease detection. I labeled features such as cell boundaries and anomalies, adhering to strict scientific protocols. My attention to detail and understanding of biological terminology ensured the dataset met rigorous quality standards, contributing to a model with 92% accuracy in identifying cancerous cells.

2022 - 2023

Education

Work History