Ahmad Faour

Senior AI / Software Engineer

Senior-leaning AI Engineer with ~4 years of combined professional experience delivering LLM-powered systems, autonomous AI agents, and production-grade services end-to-end — from data preparation and fine-tuning (PEFT/LoRA, RLHF) to optimization, ONNX export, and cloud deployment on AWS, GCP, and IIS.

  • Bridges Python ML stacks with ASP.NET Core back ends and React front ends.
  • Hands-on with agentic patterns (tool use, planning/execution, vector memory).
  • Orchestrates workflows with n8n and Elsa Workflows, including human-in-the-loop flows.
View AI Projects Contact Me
LLMs & Agents MLOps & Deployment Full-Stack AI Apps
Portrait of Ahmad Faour, Senior AI / Software Engineer

About Me

Ahmad working on AI and software engineering projects

Building real-world AI systems, not just models

I’m a Senior-leaning AI / Software Engineer with experience spanning AI/ML engineering, data science, and full-stack development. I design and ship LLM-powered applications, autonomous AI agents, and production-grade services — from data and training pipelines to APIs, front ends, and CI/CD.

My work bridges Python ML ecosystems with ASP.NET Core back ends, React UIs, and cloud infrastructure (AWS, GCP, IIS). I’m particularly focused on agentic patterns, retrieval workflows, quantization and optimization, and robust deployment of AI systems in production.

Location
Riyadh, Saudi Arabia (open to remote/global roles)
Focus Areas
LLMs, AI agents, RAG pipelines, computer vision, NLP, MLOps & deployment.
Experience
NVSSoft, Reality AI Lab, Darrebni, Freelancer.com, CampWallah.
Education
M.Sc. Artificial Intelligence (in progress) — University of Hull
B.Sc. Information Technology Engineering — University of Kalamoon (2nd in class)

Professional Experience

Dec 2024 – Present · Riyadh, Saudi Arabia

AI Engineer (Full-time, On-site)

NVSSoft
  • Designed and implemented an ASP.NET Core application with a secure REST API layer (Entity Framework/DbContext) and Python microservices for model inference.
  • Configured IIS and production environments for high availability and throughput; partnered with DevOps to enable CI/CD with automated builds, tests, and multi-stage deployments.
  • Built a React front end integrated with the .NET back end for real-time AI interactions, reducing response latency and improving UX.
  • Instrumented AI agent workflows using n8n and Elsa Workflows (webhooks, scheduled jobs, tool chaining, retries, and human-in-the-loop approvals).
  • Applied SOLID principles and clean architecture; authored internal documentation and onboarding runbooks.
Dec 2024 – Present · Remote

AI Engineer (Part-time, Remote)

Reality AI Lab
  • Contributed to Marvel Model, a next-gen generative image model, building Python training/fine-tuning pipelines to improve stability and scalability.
  • Developed tooling for data augmentation, feature extraction, and custom metrics for image-based evaluation; presented progress in sprint reviews.
Sep 2024 – Present · Remote

Data Scientist Intern (Part-time, Remote)

Darrebni
  • Built predictive models for customer behavior and market trends using advanced SQL and ML, running uni/bivariate/multivariate analyses for actionable insights.
  • Performed feature engineering (preprocessing, normalization, dimensionality reduction) and anomaly detection to improve decision quality.
  • Created interactive Tableau dashboards and collaborated with engineering/product teams to integrate insights into workflows.
Sep 2022 – Dec 2024 · Remote

AI Engineer (Part-time, Remote)

Freelancer.com
  • Delivered NLP solutions (classification, summarization, sentiment analysis) using Transformers; built Arabic handwriting recognition to ~85% accuracy.
  • Applied quantization and ONNX export to reduce model size by 59% and accelerate inference on CPU-only environments (e.g. PythonAnywhere).
  • Constructed agentic/NLP workflows with LangChain, fine-tuning models with RLHF and PEFT to improve response quality.
  • Optimized Node.js/Express back ends, improving throughput and reliability by ~20%.
Apr 2024 – Jun 2024 · Remote

Machine Learning Intern (Part-time, Remote)

CampWallah Sholay, India
  • Integrated ML models into production, increasing accuracy by ~25% through improved pipelines for real-time inference and secure API delivery.
  • Contributed to API design for seamless deployment and monitoring of ML models.

Technical Skills

Programming

Python, C#, JavaScript (React, Node.js), SQL, MATLAB

ML & Deep Learning

TensorFlow, PyTorch, Keras, scikit-learn, Pandas, NumPy; computer vision & NLP pipelines.

LLMs & Agents

Hugging Face Transformers, LangChain, RAG, tool-calling, planner/executor patterns, vector memory, human-in-the-loop.

MLOps & Deployment

Quantization, ONNX, Docker, REST APIs, ASP.NET Core, CI/CD, observability, IIS, PythonAnywhere, Heroku.

Orchestration & Workflows

n8n, Elsa Workflows, scheduled jobs, retries, event-driven flows, approvals and HIL.

Soft Skills

Analytical thinking, problem solving, collaboration, communication, adaptability, time management, research, attention to detail, troubleshooting.

Selected Projects

Smart Image Captioning with MoE

Mixture-of-Experts Computer Vision LLMs Streamlit

Reality AI Lab · Sep 2025 – Oct 2025

  • Built a Mixture-of-Experts system combining a ResNet-50 vision expert with LLM-based captioning (via Ollama, e.g., llama3) using a gating network for confidence-aware routing.
  • Designed modular moe/ components (classifier, gating, templates, LLM client) with YAML-based configs and rich logging.
  • Delivered both Streamlit UI for interactive testing and a CLI for batch captioning; packaged for Dockerized deployment.
View on GitHub Vision · MoE · LLMs

Schema-Aware Query Generator

RAG SQL Streamlit Vanna

Personal · Mar 2025 – May 2025

  • Built a retrieval-augmented query assistant that ingests SQL DDL, docs, and prior queries, indexes them in ChromaDB, and uses Vanna+GPT to generate executable SQL from natural language.
  • Implemented robust MSSQL connectivity and automatic DDL chunking for large schemas, plus incremental training from user uploads.
  • Added a safe query runner (no SELECT *) and multi-chart analytics to accelerate ad-hoc reporting and reduce syntax errors.
View on GitHub RAG · SQL · Analytics

AI Assistant with Voice Communication

Whisper v3 Arabic NLP React Flask

Freelancer.com · Oct 2024 – Dec 2024

  • Architected a multilingual, dialect-aware AI voice assistant using Whisper v3 and a fine-tuned UBC-NLP/MARBERTv2 model for Arabic dialect classification, with GPT-family LLMs for reasoning.
  • Designed a modular, sub-500ms pipeline integrating ASR, NLP, and reasoning modules via REST.
  • Built Flask API backend + React UI (WebSockets) and orchestrated flows in n8n/Elsa for events, retries, and human-in-the-loop review.
Voice · LLMs · Orchestration

Arabic Handwriting Recognition

TensorFlow ONNX Flask

Personal / Freelance · Jun 2024 – Jul 2024

  • Developed a CNN-based Arabic character recognition model trained on 120k handwritten samples across 28 classes, achieving 85% top-1 accuracy.
  • Applied quantization-aware training and ONNX conversion, reducing model size by 59% and achieving 2.3x CPU inference speedup.
  • Deployed via Flask on PythonAnywhere, serving 5k+ requests/day without downtime.
CV · Quantization · Deployment

Education & Certifications

Education

University of Hull

M.Sc. in Artificial Intelligence (in progress)
Sep 2025 – Present · Hull, United Kingdom
Focus on computer vision, deep learning, and LLM applications.

University of Kalamoon

B.Sc. in Information Technology Engineering
GPA: 3.08 (83/100) · 2nd highest rank in class
Aug 2018 – Feb 2024 · Deiratiah, Syria
Coursework: AI, ML, Neural Networks, Data Mining, Pattern Recognition, Deep Learning, NLP, Algorithms, Database Systems, Cybersecurity.

Yousef Seman Industrial High School

Diploma in Computer Technology
Aug 2016 – May 2018 · Yabroud, Syria
GPA 92/100; ranked 5th in governorate. Courses in programming, electronics, digital logic, networking, and applied mathematics.

Certifications & Awards

Generative AI with Large Language Models

Coursera & AWS · Oct 2024
Mastered transformer architectures, PEFT/LoRA, and RLHF; deployed LLMs on cloud.

Introduction to Retrieval Augmented Generation (RAG)

Duke University & Coursera · Dec 2024
Built RAG pipelines using Python, LLMs, and vector databases.

Intermediate Machine Learning

Kaggle · Feb 2025
Applied XGBoost, advanced preprocessing, and cross-validation.

Feature Engineering

Kaggle · Nov 2024
Designed feature engineering workflows to boost model performance.

AWS EMEA Innovate: Migrate. Modernize. Build.

AWS · Oct 2024
Learned scalable AI deployments on AWS cloud infrastructure.

Azure DevOps: Intro to CI/CD

United Latino Students Association · Mar 2024
Integrated ML model deployment into CI/CD workflows.

Data Science Internship

Darrebni · Mar 2025
Predictive modeling, data visualization, and SQL pipelines supporting business decisions.

Get In Touch

Contact Information

Feel free to reach out to discuss AI projects, collaboration opportunities, or just to connect!

Location

Riyadh, Saudi Arabia

Email

ahmad.saleh.faour@gmail.com

Phone

+966-596824729

LinkedIn

linkedin.com/in/ahmad-faour

GitHub

github.com/ahmadfaour9