Abdourahman Djamal Djamal

Key Skills

Software

CVAT

Labelbox

Top Subject Matter

No subject matter listed

Top Data Types

Image

Computer Code Programming

Top Label Types

Bounding Box

Polygon

RLHF

Computer Programming Coding

Freelancer Overview

I have hands-on experience applying AI and machine learning to cybersecurity, where I have designed and implemented anomaly detection systems and developed Python automation scripts for data processing and log analysis. My work includes building and optimizing machine learning models for identifying abnormal network traffic, which required precise data labeling and feature engineering to ensure high detection accuracy and minimize false positives. I am adept at preparing, annotating, and validating large datasets for supervised and unsupervised learning, and have a strong background in using Python, SQL, and SIEM tools (Splunk, ELK, Wazuh) for data-driven security solutions. My projects have involved end-to-end data pipeline development, from data collection and preprocessing to model training and evaluation, making me well-versed in the full lifecycle of AI training data preparation.

Labeling Experience

Project: Python Code Optimization & Functional Correctness Annotation

LabelboxComputer Code ProgrammingRLHFComputer Programming Coding

Annotated a dataset of 2,000+ Python snippets for an LLM training project. My role involved evaluating model-generated code for logical soundness, PEP 8 compliance, and algorithmic efficiency. Specific tasks included: Identifying 'hallucinated' libraries or non-existent methods. Categorizing code by complexity (O(n) vs O(n²)). Debugging syntax errors in snippets involving Data Science libraries (NumPy, Pandas). Ranking multiple model outputs based on readability and execution speed. Maintained a 97% internal audit score by ensuring all labels strictly followed the project’s security and performance guidelines.

Annotated a dataset of 2,000+ Python snippets for an LLM training project. My role involved evaluating model-generated code for logical soundness, PEP 8 compliance, and algorithmic efficiency. Specific tasks included: Identifying 'hallucinated' libraries or non-existent methods. Categorizing code by complexity (O(n) vs O(n²)). Debugging syntax errors in snippets involving Data Science libraries (NumPy, Pandas). Ranking multiple model outputs based on readability and execution speed. Maintained a 97% internal audit score by ensuring all labels strictly followed the project’s security and performance guidelines.

2025 - 2025

Project Title: AI-Powered Visual Data Enrichment Solution for E-Commerce

CVATImageBounding BoxPolygon

Project Title: AI-Powered Visual Data Enrichment Solution for E-Commerce Description: This project delivers a custom AI solution designed to solve the business challenge of inefficient product catalog management. The primary goal is to automate the generation of rich, accurate metadata for thousands of product images, thereby improving product discoverability and reducing manual labor costs.

Project Title: AI-Powered Visual Data Enrichment Solution for E-Commerce Description: This project delivers a custom AI solution designed to solve the business challenge of inefficient product catalog management. The primary goal is to automate the generation of rich, accurate metadata for thousands of product images, thereby improving product discoverability and reducing manual labor costs.

2025 - 2025

Education

Work History