Mohamed Mghezzi Bekhouche

Key Skills

Software

Axiom AI

Google Cloud Vertex AI

Other

Top Subject Matter

Technical Document Information Extraction for Aerospace Engineering

Intent and Sentiment Classification for Customer Chatbot

Top Data Types

Text

Document

Top Label Types

Prompt Response Writing SFT

Entity Ner Classification

Classification

Freelancer Overview

Ingenieur IA (AI Engineer): Information extraction using LLM RAG systems. Brings 3+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Google Cloud Vertex AI and Other. Education includes Master of Science, CY Cergy Paris Université, ENSEA Cergy (2024) and Master of Science, Université Mohamed Khider Biskra (2023). AI-training focus includes data types such as Document and Text and labeling workflows including Entity (NER) Classification and Classification.

Labeling Experience

Ingenieur IA (AI Engineer): Information extraction using LLM RAG systems

Google Cloud Vertex AIDocumentEntity Ner Classification

Developed a solution using LLMs with LangChain RAG to extract technical information from complex PDF documents. Labeled and annotated key entities such as torque values and engineering parameters from technical documents to support AI-driven assistance for assembly engineers. Fine-tuned prompt engineering methods for extracting structured knowledge from unstructured inputs. • Utilized document and table extraction tools like Tabula and pdf-plumber for annotation. • Engineered two-stage-prompting for improved information extraction accuracy. • Focused on assisting aircraft assembly processes through AI-driven insights. • Leveraged Google Cloud Platform and Vertex AI for data processing and model deployment.

Developed a solution using LLMs with LangChain RAG to extract technical information from complex PDF documents. Labeled and annotated key entities such as torque values and engineering parameters from technical documents to support AI-driven assistance for assembly engineers. Fine-tuned prompt engineering methods for extracting structured knowledge from unstructured inputs. • Utilized document and table extraction tools like Tabula and pdf-plumber for annotation. • Engineered two-stage-prompting for improved information extraction accuracy. • Focused on assisting aircraft assembly processes through AI-driven insights. • Leveraged Google Cloud Platform and Vertex AI for data processing and model deployment.

2024 - 2024

Master's Internship: Data labeling for intent and sentiment classification

OtherTextClassification

Developed and labeled datasets for chatbot intent and sentiment prediction as part of a Master's thesis. Labeled intent categories and annotated sentiment using deep learning models including BERT and Glove-LSTM. Fine-tuned and validated classification models for accurate customer intent and sentiment analysis. • Labeled customer request texts and annotated sentiment labels. • Used BERT for intent classification and Glove-LSTM for sentiment annotation. • Created gold-standard datasets for chatbot training and evaluation. • Designed and conducted experiments for model validation and improvement.

Developed and labeled datasets for chatbot intent and sentiment prediction as part of a Master's thesis. Labeled intent categories and annotated sentiment using deep learning models including BERT and Glove-LSTM. Fine-tuned and validated classification models for accurate customer intent and sentiment analysis. • Labeled customer request texts and annotated sentiment labels. • Used BERT for intent classification and Glove-LSTM for sentiment annotation. • Created gold-standard datasets for chatbot training and evaluation. • Designed and conducted experiments for model validation and improvement.

2023 - 2023

Education

Work History