Looking for a job

khadija bayoud

@khadijabayoud

Message

AI Engineer at ToumAI building Darija speech-to-text models and voice analytics for customer experience.

Morocco

Message

At ToumAI, I build speech-to-text models for Moroccan and Tunisian Darija, along with on-premise pipelines for voice privacy and PII anonymization.

I delivered the analytics layer of HolisticX, which analyzes enterprise call centers and virtual voice agents to surface business insights, agent performance trends, and customer preferences. I also delivered a Microsoft-certified SaaS solution connecting multilingual AI assistants with Microsoft Dynamics 365.

Previously, at DeepEcho, I researched adapting vision-language models to fetal ultrasound data and implemented an image classification pipeline. My earlier internships included developing an automated media content generation platform at L’ODJ Média and a Laravel application for aircraft component documentation at SAFRAN Aerosystems.

Experience

Work history, roles, and key accomplishments

Current

AI Engineer

Current

ToumAI

Dec 2024 - Present (1 year 10 months)

Building speech-to-text models for multilingual dialects, on-premise data-privacy pipelines, and reproducible ML workflows. Delivered speech-to-text and NER models for Darija and Tunisian, and the analytics layer for the HolisticX product.

Speech to Text NLP Anonymization Data Privacy ML Pipelines Microservices Event Driven Architecture OpenAI Whisper Qwen ASR NVIDIA NeMo BERT GLiNER Speaker Diarization Emotion Detection Microsoft Dynamics 365

Data Scientist

DeepEcho

Feb 2024 - Nov 2024 (9 months)

Conducted research on adapting Vision-Language Models to fetal ultrasound data through parameter-efficient fine-tuning. Implemented an image classification pipeline to categorize fetal ultrasound images by type.

Vision Language Models CLIP BLIP 2 LLaVA LoRAs Fine Tuning Model Evaluation BLEU ROUGE BERTScore F1 Score Image Classification

Internship

L'ODJ Média

Jul 2023 - Aug 2023 (1 month)

Developed an All-in-One automated media content generation platform integrating news scraping, LLMs for article generation, TTS models for voice synthesis, and diffusion models for image/video generation.

News Scraping LLM TTS Diffusion Models Content Generation Human In The Loop

Internship

SAFRAN Aerosystems

Jul 2022 - Oct 2022 (3 months)

Implemented a full-stack web application with Laravel, featuring front-end components, RESTful back-end services, and a MySQL database to digitize and manage aircraft component documentation and reporting workflows.

Laravel Fullstack Development REST APIs MySQL Web Application Development