~/portfolio (main) $

Machine Learning,
Computer Vision,
And Photography.
Spotting the beauty in the little things.

// About me
const me = {
  role: "AI Engineering Student",
  passion: ["computer vision", "photography"],
  currently: seekingPhD(
    "AI", "ML", "RL",
    "Computer Vision", "Optimization")
}

View Projects → Photo Gallery Latest Posts
Always
Learning 🌱
Profile photo
∞
Love Life
02

AI Projects

001

Generative Adversarial Networks (GANs)

Implemented GAN architectures for image generation and representation learning, exploring training stability and generative quality.

GANGenerative AIDeep Learning
002

Diffusion Models for Text Generation

Exploring diffusion-based generative models applied to text, studying their capabilities as alternatives to autoregressive LLMs.

DiffusionGenerative AINLP
003

LLM-Based Legal Explanation Agent

Designed a RAG-based agent for French legal procedures using AWS Bedrock, SageMaker, FAISS, and LangChain, enabling contextual and explainable legal assistance.

LLMRAGAWSLangChainFAISS
→
004

Multimodal Video Analysis with VLM

Built a multimodal system using Vision-Language Models (InternVL3.5-1B) for real-time video and webcam understanding with interactive Streamlit interface.

VLMMultimodalComputer VisionStreamlit
→
005

LEA-UNet — Medical Image Segmentation

Reproduced LEA-UNet (Ouyang et al., 2023) with attention mechanisms for retinal vessel segmentation, scaling datasets via augmentation and evaluating with Dice & IoU.

Deep LearningMedical ImagingCNNSegmentation
006

Neural Geometric Digital Twin of Airports

Exploring neural implicit representations (DeepSDF) for 3D scene modeling, building pipelines from OpenStreetMap to signed distance fields with MLPs and Fourier features.

3D AIDeepSDFNeural FieldsMLP
007

Big Data Pipeline for Urban Mobility

Designed an end-to-end pipeline on NYC taxi data from ingestion to analytics, with ML prediction of trip costs, using Airflow and Streamlit dashboards.

Big DataAirflowMLData Engineering
→
008

EV Charging Optimization (Research Study)

Conducting a state-of-the-art study on AI-based optimization for EV charging using multi-agent systems and combinatorial optimization approaches.

ResearchOptimizationMulti-AgentEnergy AI
03

Latest Posts

From 2D Maps to 3D Models: Building Water-Tight Meshes from GeoPackage Data

A beginner-friendly guide to transforming flat airport footprints into solid 3D geometry — the first step toward an AI-powered Digital Twin.

See post →

Coming Soon

Yeah, I know you are curious, buuuuuuuuut you need to wait a bit till I publish my next post! 😉.

04

About Me

I'm a Computer Engineering student specializing in Artificial Intelligence, building systems that understand the visual world while I explore it with my camera.

My Motto?
"Love your Life & Find beauty in the little things!"

Get in touch →
🧠
PyTorch
👁️
Computer Vision
🌐
Machine Learning
💬
NLP / LLMs
⚡
Python
📷
Photography

Let's connect

Have an idea?
Let's build it.

zaynab.merimi.2@email.com