Mohammed AlrehailiAI Engineer
A practice in applied AIPortfolio / 01

Intelligence,in practice.

I build AI systems that turn complex information into practical decisions.

The approachI / IV
From MODELS
to MEANINGFUL
DECISIONS.
Engineering with a purpose.About me
A strong foundation

4.61/ 5

Computer Science
Graduated with honors.

03AeroSeed
Concept illustration
Best AI Project Award30+ students mentoredAzure AI certified
Explore the work

01 / Selected work

Ideas, made real.

Four different challenges.
One focus: AI with a practical purpose.

AgroMind / Learning from nature LEAF FEATURES
Vision / Transfer learningConcept illustration
02May – Jun 2026

Plant disease detection & MLOps

AgroMind

A closer look at plant health.

A leaf image becomes a practical next step. AgroMind brings EfficientNet disease classification, tracked model experiments, and mitigation guidance into a Streamlit application.

Input
Uploaded photographs of crop leaves.
Output
Disease classifications and recommended mitigation strategies.
  • PyTorch
  • EfficientNet
  • MLflow
  • Streamlit
  • Transfer Learning
Explore the model & MLOps: AgroMind

From experiments to inference

  1. Fine-tune the vision model

    PyTorch and an EfficientNet CNN backbone extract visual features from leaf imagery. Transfer learning adapts the model to agricultural disease categories.

  2. Track and compare

    MLflow records hyperparameters, loss curves, checkpoints, and evaluation metrics. More than 20 training runs support comparison and checkpoint selection.

  3. Make predictions accessible

    A Streamlit interface accepts leaf photos and presents disease classifications alongside recommended mitigation strategies.

Model design

EfficientNet transfer learning balances image classification with a lightweight inference footprint.

Experiment visibility

Training settings, model artifacts, and evaluation results are tracked together in MLflow to make model iterations easier to inspect.

The deliverable

A plant disease detection application with an experiment-tracking workflow behind it and a simple upload experience in front.

AeroSeed / Atmospheric intelligence SEEDING WINDOW
Weather patterns / Predictive MLConcept illustration
03Oct – Dec 2024

Cloud seeding prediction & decision support

AeroSeed

Finding possibility in the sky.

AeroSeed predicts the precipitation increase from cloud seeding using weather, location, and seeding information—helping experts choose when and where to seed.

Input
Weather conditions, location, and cloud seeding information.
Output
Predicted precipitation increase to support timing and location decisions.
  • Python
  • Machine Learning
  • Scikit-learn

Best AI Project Award
6th Place · National HPC/AI Hackathon

Explore the prediction workflow: AeroSeed

From conditions to opportunity

  1. Bring the context together

    Use weather conditions, geographic location, and seeding information as inputs to the prediction workflow.

  2. Estimate precipitation increase

    Apply machine learning to predict the additional precipitation associated with cloud seeding under those conditions.

  3. Support expert decisions

    Use the predictions to inform the timing and location of cloud seeding operations.

The purpose

Translate weather and seeding data into actionable predictions for experts planning cloud seeding operations.

Recognition

Best AI Project Award at the Technical & Artificial Intelligence Talents Forum, and 6th Place at the National HPC/AI Hackathon.

02 / The person behind the work

Curious by nature.
Practical by design.

I’m Mohammed, an AI Engineer and Computer Science honors graduate. I work across Generative AI, RAG, and computer vision, connecting models with the backends and interfaces that make them useful.

My work spans architectural auditing, agricultural diagnostics, atmospheric prediction, and motorsport analytics. I also enjoy making technical ideas accessible—an approach I brought to mentoring more than 30 students in AI and robotics.

Mohammed AlrehailiAlways learning.
Always building.
01

Generative AI

LLMs · AI agents · RAG architecture · Vector search · LangChain

02

Deep learning & vision

PyTorch · EfficientNet · CNNs · YOLO · OpenCV

03

Delivery & MLOps

FastAPI · Streamlit · MLflow · Docker · Git & GitHub · REST APIs · Microsoft Azure · Azure AI Services · Azure OpenAI

04

Data & programming

Python · SQL · C · Java · Scikit-learn · XGBoost · Random Forest · Logistic Regression · NumPy · Pandas

03 / Experience & learning

Built on experience.

Hands-on fellowships, thoughtful experimentation, and knowledge shared.

May – Jun 2026Riyadh, Saudi Arabia · Remote

Saudi Digital Academy
& WeCloudData

AI Engineer Fellow

Built GenAI and RAG pipelines with vector search and LangChain agents. Developed PyTorch workflows tracked in MLflow, FastAPI inference endpoints, and Streamlit dashboards.

Nov 2025 – Jan 2026Riyadh, Saudi Arabia

Tuwaiq Academy

AI Model Building & Development Fellow

Designed and fine-tuned computer vision models for visual inspection. Connected multimodal LLMs with Python backends to turn visual outputs into structured decisions.

Jul – Sep 2025Medina, Saudi Arabia

Valley Taibah Academy

AI Specialist Intern

Mentored 30+ students in AI and robotics. Guided practical Python projects, hardware/software integration, debugging, and capstone demonstrations.

The foundation / Education

Islamic University

Bachelor of Science in Computer Science

4.61/ 5.00 · Honors

Jul 2020 – Sep 2025
Medina, Saudi Arabia

Continuing the practice / Credentials

Microsoft Certified: Azure AI Apps and Agents Developer Associate

Microsoft · AI-103

Jul 2026

Microsoft Certified: Azure AI Engineer Associate

Microsoft · AI-102

Oct 2024

Program in Artificial Intelligence

KAUST

Sep 2023

The full picture.

Experience, projects, and credentials in one PDF.

Preview the CV here

Use “Open CV” above to view the document. An inline preview loads here when JavaScript and PDF viewing are supported.

04 / Start a conversationA good place to begin.

Let’s build
something
useful.

Have a role, a project, or an interesting AI problem in mind? I’d love to hear about it.