Maryam Maksour

    AI Engineer · Dubai, UAE
  

Maryam
Maksour

AI Engineer Agentic Systems · LLMs · RAG · Arabic NLP

Building production-grade AI systems at the intersection of language, data, and healthcare — with a background in competitive programming that shapes every problem I solve.

Download CV Get in touch

      GitHub
      LinkedIn
      Email
    

Background

Education

Bachelor of Science — Information Technology

2019 – 2024

Arab Private University for Science and Technology, Hama, Syria

→GPA 3.59 / 4.00 — ranked 2nd in cohort
→Consistently ranked among the top 3 students in the program throughout all four years
→Graduation Project: Integrated ChatGPT into a mobile application designed to help Information Technology / Informatics Engineering students prepare for Syria's national licensing examination

Career

Experience

Aug 2025 – Mar 2026

Evolutions

AI Engineer

·Built a production-grade agentic AI system: a ReAct-based AI agent using vLLM for high-throughput local inference
·Orchestrated multi-step reasoning workflows with LangGraph state machines
·Designed a modular tool architecture enabling autonomous task decomposition and sequential tool chaining

Jan 2025 – Mar 2025

Al Dar Academy

Instructor — Big Data & SDLC

·Taught Big Data & Visualization and Software Development Life Cycle to undergraduate students
·Designed an interactive curriculum focused on real-world application of technical concepts

Mar 2024 – Jun 2024

SMSM AI

Machine Learning Intern

·Preprocessed and analyzed audio data for fine-tuning Whisper models

2020 – 2021

RightClick Company

AI Engineer Intern

·Completed AI engineering training as the youngest engineer at the company
·Contributed to real-world AI projects alongside senior engineers

Work

Projects

      Featured
    

Hybrid RAG + Text-to-SQL System for Mixed-Type Data

Designed a hybrid architecture for AI chatbots over knowledge sources containing both numeric and free-text columns — routing each column to the retrieval method best suited to its content. Numeric columns are handled via SQL querying; free-text columns via semantic search — combined within a single query pipeline. Implemented on PostgreSQL.

RAG Text-to-SQL PostgreSQL LLM Semantic Search

Arabic ASR Data Research — Egyptian Dialect

Surveyed major datasets for Egyptian Arabic ASR (Common Voice, MGB-2/3, FLEURS, CALLHOME, ArZen, and others), comparing WER results across existing models to identify high-performing data sources. Contributed to validating newly collected data for the project.

        Arabic NLP·ASR·Whisper·Research
      

PDF Intelligent Assistant

RAG-based chatbot for PDF question-answering using advanced LLMs. Allows users to upload documents and query their content conversationally.

        RAG·LLM·Chatbot
      

AI Customer Service Chatbot

OpenAI-powered automated response system built with prompt engineering for consistent, context-aware customer interactions.

        OpenAI API·Prompt Engineering·Chatbot
      

Audio Classification: AST vs. HuBERT

Fine-tuned and benchmarked Audio Spectrogram Transformer and HuBERT on the Speech Commands dataset to compare model performance on audio classification.

        HuggingFace·Audio ML·Transformers
      

AI Personal Trainer

Real-time workout form analysis and repetition counting using computer vision and pose estimation.

        Computer Vision·OpenCV·Real-time
      

Gesture Volume Control

Hand landmark detection for system-level volume control, using computer vision to translate gestures into OS-level audio commands.

        Computer Vision·MediaPipe·OpenCV
      

Publication

Research & Publication

“Comparison of Different GPT Versions and the Effectiveness of ChatGPT in the Medical Sector”

Journal of Hama University, Syria · Co-authored with Dr. Maisaa Diab 2024

01Reviewed the evolution of GPT models from GPT-1 through GPT-4 / ChatGPT, tracing improvements in scale, alignment, and capability across versions

02Studied the social and technical impact of ChatGPT adoption in healthcare through a survey of 93 respondents from a range of medical backgrounds

03Findings: high perceived reliability (mean 4.19 / 5) and ease of use (mean 4.11 / 5) among respondents

04Concluded with recommendations for developers, including enriching Arabic-language medical training data

Competitor & Coach

Competitive Programming

As a Competitor

2nd

Syrian Private Universities Programming Collegiate Contest

2023 — team competition

3rd

Syrian Private Universities Programming Collegiate Contest

2024 — team competition

8th

ACPC Girls Division — Middle East Women's Programming Contest

2023 — online

As a Coach

Competitive Programming Coach — Arab Private University for Science and Technology, 2024

Coached alongside multiple co-coaches, leading 5 student teams in preparation for collegiate contests

All five teams achieved strong results — several advanced to SCPCP, one team qualified for the Arab Collegiate Programming Contest (ACPC)

Mentored students in critical thinking, problem-solving strategy, and team dynamics

Technical

Skills

Languages

Python (advanced) C++ (advanced) SQL R (basics)

ML / AI

TensorFlow Scikit-learn HuggingFace LangChain LangGraph vLLM

Data

NumPy Pandas Matplotlib Seaborn OpenCV PySpark

NLP

Transformers Whisper fine-tuning Arabic NLP RAG pipelines

APIs & Cloud

FastAPI REST API AWS OpenAI API

Databases

PostgreSQL

Community

Volunteering

Python Tutor & Mentor

Oct 2023 – May 2025

Paper Airplanes

Mentored students online; developed structured teaching materials and communication skills.

Say hello

Contact

I'm open to new opportunities, research collaborations, and interesting conversations. Reach out via any of the channels below.

Email

maryammaksour@gmail.com

LinkedIn

linkedin.com/in/maryam-maksour

GitHub

github.com/MaryamMaksour

Location

Dubai, UAE