Ahmed H.

Result-oriented Electrical Engineer and Arabic Language & AI Specialist with extensive hands-on experience in AI model training, Reinforcement Learning from Human Feedback (RLHF), audio transcription, and data annotation. Over the years, I have logged extensive freelance hours across platforms like Upwork and with specialized AI data vendors (including Appen and Neuro.net), delivering high-quality speech and text datasets. My work ranges from complex verbatim audio transcription and voice annotation for IVR systems to evaluating LLM responses in Modern Standard Arabic (MSA) and various regional dialects. By combining the analytical rigor of an engineering background with deep linguistic precision, I help AI enterprises benchmark and optimize next-generation language and voice models. Core Competencies: AI Model Evaluation & RLHF: Ranking, editing, and evaluating AI-generated responses for factual accuracy, linguistic fluency, grammar, and cultural relevance. Audio Transcription & Annotation: 100% verbatim transcription, speaker tagging, timestamping, and acoustic event labeling (crosstalk, fillers, background noise). Voice & IVR Systems: Audio data collection, evaluation, and automated Interactive Voice Response (IVR) dataset creation (including major telecom projects like Zain KSA). Dialectal Expertise: Native proficiency in MSA, Saudi (Hijazi), and Egyptian Arabic, alongside deep comprehension of Levantine (Syrian/Lebanese) and Iraqi dialects. Technical & Analytical Foundation: B.Sc. in Electrical Engineering, advanced mathematics (Calculus & Linear Algebra), Python syntax, and C programming. Open to remote opportunities in AI Data Annotation, LLM Evaluation, Arabic Language Quality Assurance, and Speech Recognition projects.