AI Engineer @ AEMS.ai | Turning Data into Impactful Solutions
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Hi there! I am an AI engineer and data enthusiast passionate about solving global challenges through data-driven insights and AI, with a particular focus on large language models (LLMs). My diverse background, multilingualism, and strong communication skills empower me to effectively express my ideas and drive cross-cultural collaboration. I am confident that any clearly defined problem can be solved with enough data. Don't believe me? Let's connect and find out!
American University of Beirut
Bachelor of Science - BS, Computer Science
August 1, 2022 – May 1, 2025
AEMS.ai
AI Research Engineer
February 1, 2026 – Present
London Area, United Kingdom · Remote
Deloitte
AI Engineer
August 1, 2025 – February 1, 2026
Dubai, United Arab Emirates · Remote
961
AI Engineer
May 1, 2025 – August 1, 2025
Kudwa
AI Engineer
January 1, 2025 – February 1, 2025
San Francisco, California, United States · Remote
Synopsys Inc
ML Engineer Intern
June 1, 2024 – July 1, 2024
Yerevan, Armenia · On-site
American University of Beirut
Competitive Programming Coordinator
September 1, 2023 – October 1, 2024
Sparse Matrix-Sparse Matrix Multiplication Optimization on GPUs
May 1, 2025 – June 1, 2025
Standard matrix multiplication algorithms are highly optimized, but they perform poorly when dealing with sparse matrices—matrices with mostly zero entries. In such cases, significant time, memory, and compute are wasted on operations involving zeros. We developed optimized algorithms tailored for sparse matrices, reducing redundant computations and memory access and achieving up to 1000× speedups compared to standard algorithms.
Detect X
April 1, 2025 – May 1, 2025
Built a scalable application that extracts actionable insights from unstructured documents by combining computer vision and NLP techniques. Deployed production-ready AI services using the AWS ML stack—including Rekognition and Comprehend—within a serverless architecture for real-time processing. 🔹 Automated extraction of key data from scanned and digital documents 🔹 Enabled real-time document analysis at scale with minimal infrastructure overhead 🔹 Integrated CV + NLP pipelines for structured understanding of unstructured content Designed for high-throughput environments where fast, reliable document intelligence is critical.
Space-Time Aware Text Embeddings
January 1, 2025 – June 1, 2025
"Two entities that are close semantically but not close in space-time, should not be as close in the embedding space. In other words, this relationship should be explicitly preserved." We finetuned a transformer based sentence embedding model after adjusting its internal architecture to take in extra inputs (space & time). We tested multiple approaches one of which was adjusting the internal attention mechanism. We show promising results that should be scaled to confirm application across larger space-time landscapes.
UDP Simulation
October 1, 2024 – November 1, 2024
A Java-based tool that simulates both reliable and unreliable network conditions over UDP. This includes packet loss, delay, and out-of-order delivery. Supports configurable loss patterns (standard, bursty, Gilbert-Elliott/Markov) and multiple delay distributions (uniform, Gaussian, exponential, triangular). ✅ Emulates real-world network behaviors ✅ Tracks detailed delivery stats (loss rate, avg delay, out-of-order count) ✅ Fully configurable via command-line ✅ Thread-safe with controlled synchronization mechanisms
AI-Driven Mental Health Assistant
April 1, 2024 – April 1, 2024
Built an AI-driven mental health screening tool during a healthcare AI hackathon to identify signs of mental illness from user speech input. Used LLAMA and LangChain for real-time sentiment analysis and dynamic mental state evaluation, with workflow coordination via Crew AI. 🔹 Integrated user responses with clinical depression indicators through an intuitive interface 🔹 Delivered instant feedback to support scalable, AI-powered pre-screening 🔹 Achieved 90%+ accuracy across diverse datasets, demonstrating strong generalizability 🔹 Designed for real-world clinical impact in early mental health assessment Focused on leveraging cutting-edge LLM frameworks to enhance emotional state recognition and support clinicians in improving patient care.
Cultural Fit Analysis
The candidate's project portfolio demonstrates a strong interest in cutting-edge AI/ML technologies and a proactive approach to learning and application. The diversity of projects, from academic research (Space-Time Embeddings, Sparse Matrix Optimization) to practical applications (Detect X, AI-Driven Mental Health Assistant), indicates a curious and driven individual. The competitive programming background suggests a collaborative and problem-solving mindset. The target role of ML Engineer aligns well with the candidate's demonstrated skills and project focus. However, the candidate's experience level (2) is relatively junior for a 'senior' evaluation, and the lack of detailed company-specific experience beyond internships makes a deep cultural fit assessment challenging.
Soft Skills & Operational Fit
The candidate's involvement in competitive programming coordination suggests leadership, mentorship, and community-building skills. Project descriptions indicate an ability to work on complex problems and articulate solutions. The diverse range of projects, from low-level optimization to high-level AI applications, suggests adaptability and a broad interest in technical challenges. However, without specific psychometric or English test results, a detailed assessment of work attitude, stress handling, and team collaboration is not possible.