AI Engineer with 1+ years in RAG Architectures & LLM Fine-Tuning
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Computer Science graduate with an academic specialization in Artificial Intelligence and hands-on experience as an AI Developer. Skilled in engineering complex RAG architectures, fine-tuning open-source LLMs for domain-specific tasks, implementing advanced prompt engineering techniques, and deploying real-time voice assistants. Adept at bridging the gap between AI models and backend APIs to deliver seamless, adaptive user experiences.
AL SHAM PRIVATE UNIVERSITY
Computer Science
August 1, 2019 – June 30, 2025
Fatoorah
AI Engineer
March 1, 2025 – Present
Saudi Arabia
SMART TEACHER
August 1, 2025 – Present
An AI-powered educational platform assisting Saudi primary students (ages 6-12) via interactive chatbots and adaptive quizzes rooted in the official curriculum. Key Results or Achievements: • RAG Pipeline Optimization: Engineered and optimized the Retrieval-Augmented Generation (RAG) pipeline, boosting retrieval accuracy significantly from 20% to 85% by upgrading the embedding model and refining the data processing workflow. • LLM Fine-Tuning & Domain Adaptation: Fine-tuned open-weight models (Qwen) using Unsloth to adapt the AI's tone and pedagogical style for Saudi primary students. Successfully optimized the generation pipeline, reducing response latency by 60% for general queries while ensuring high accuracy and context-adherence for RAG-dependent queries. • Multimodal Voice Interaction: Developed a seamless voice-to-voice pipeline integrating Speech-to-Text (STT) and Text-to-Speech (TTS); enabled students to ask verbal questions and receive audio explanations, significantly enhancing accessibility for young learners. • Vision & OCR Integration: Integrated vision capabilities allowing students to upload textbook images, extract text via OCR, and receive context-aware explanations. • Advanced Prompt Engineering: Implemented Chain of Thought (COT) and Few-Shot Learning techniques alongside a query classifier to handle edge cases and improve answer precision. • Query Enhancement Module: Developed a "Question Optimizer" to handle context (Chat History) and autocorrect spelling errors, tailoring the UX for young users with limited typing skills. • Performance & Cost Efficiency: Reduced quiz generation latency by 60% and minimized token usage by optimizing logic and implementing score thresholds in the retriever mechanism. • Adaptive Learning System: Built a personalized testing engine that adapts to student performance, specifically targeting previous mistakes to reinforce learning concepts. • Backend API Development: Designed and implemented backend endpoints to support AI-driven features, enabling seamless interaction between the client applications and multiple AI services.
AI ACCOUNTANT
May 1, 2025 – July 1, 2025
An AI-powered accounting chatbot specialized in the Fatoorah platform, capable of answering accounting inquiries, guiding users through financial operations, and dynamically generating SQL queries to produce advanced analytical reports. Key Results or Achievements: • Evaluation & Benchmarking: Engineered a comprehensive testing module and curated a specialized dataset to benchmark the agent's accuracy across financial scenarios. • Agent Logic Optimization: Refined the core agentic workflow and reasoning logic, resulting in a measurable improvement in answer quality. • Advanced Context Management: Resolved critical chat history issues and implemented an optimized conversation summarization strategy, enabling the system to maintain long-term context and provide coherent multi-turn responses. • Iterative Prompt Engineering: Systematically improved system prompts and routing mechanisms based on test feedback to handle complex accounting queries effectively.
AIVoiceCall
March 1, 2025 – April 1, 2025
An AI-powered voice assistant integrated with the Fatoorah platform that answers customer accounting inquiries through natural voice interaction, providing accurate and real-time financial guidance. Key Results or Achievements: • Strategic Model Selection: Conducted a comparative cost-benefit analysis of various STT and TTS models to define the optimal technology stack, balancing audio quality with operational costs. • Secure Real-Time Architecture: Engineered a dynamic Access Token system using LiveKit with granular permissions including SIP support, secure room entry, and audio publishing ensuring robust access control and stable connectivity. • End-to-End Pipeline Integration: Developed a responsive full-stack interface linked to the backend, enabling real-time interaction with the Voice Agent and validating the data flow between system components. • Iterative Performance Optimization: Executed rigorous testing cycles on the Voice Agent, leading to a strategic migration to a superior model that enhanced response accuracy and reduced latency.
Unsupervised Learning, Recommenders, Reinforcement Learning
Coursera
December 1, 2023 – Present
AI For Everyone
Coursera
June 1, 2023 – Present
Cultural Fit Analysis
The candidate's projects demonstrate a strong alignment with the target role of an AI Engineer, focusing on practical applications of AI, particularly in conversational AI, educational technology, and financial services. The diversity of projects (educational platform, accounting chatbot, voice assistant) shows adaptability and a broad interest in applying AI across different domains. The certifications in 'AI For Everyone' and 'Unsupervised Learning, Recommenders, Reinforcement Learning' indicate a commitment to continuous learning and staying updated with AI advancements. The candidate's experience is primarily within a single company (Fatoorah) and its related projects, which might suggest a focused but potentially less diverse exposure to different organizational cultures or project methodologies outside this specific environment.
Soft Skills & Operational Fit
The candidate's project descriptions highlight problem-solving skills (e.g., boosting retrieval accuracy, reducing latency, resolving chat history issues) and an iterative approach to development (e.g., iterative prompt engineering, rigorous testing cycles). The focus on user experience for young learners and financial users indicates an ability to consider practical application and user needs. However, without direct assessment data, specific soft skills like teamwork, leadership, or stress handling cannot be definitively evaluated.