Data Scientist · RAG / LLM / NLP · Switzerland

Fatima Ezzahra
Sounny Slitine

I find the structure hidden in messy human data — building knowledge graphs, semantic clusters, and RAG & LLM systems from raw text, news, and networks.

Drag a node — or click a project to jump to it.
scroll

Node · 00 — Who

I'm a data scientist and AI developer working between Switzerland and Morocco. My work lives where language meets networks.

I extract entities and relations from text, score them on knowledge graphs, cluster people and ideas by meaning, and wrap the result in production APIs. I've mapped misinformation across a country, matched learners by their goals, and taught pipelines to turn paragraphs into video. The thread through all of it: turning noise into something you can reason about.

  • Based inSwitzerland
  • FocusRAG · NLP · Graph ML · LLMs
  • Open toAI / Data roles & collaborations

Cluster · 01 — Selected work

Things I built that found the signal

Four projects, each one a node in the graph above. Same job every time: pull structure out of unstructured human data.

node · knowledge-graph 2023 — now

Epineon.AI · Casablanca · Data Scientist · 2023—now

A platform that reads the news and scores what matters

An upskilling and networking app that scrapes news at scale, then ranks each article by the entities and relations inside it using a knowledge graph — so people see what's actually worth their time.

  • Generated personalised learning content with LLMs — summaries, text-to-speech, podcasts, and quizzes.
  • Clustered users by goal: SentenceTransformers embeddings grouped with Ward's hierarchical clustering, plus a custom threshold that finds the cleanest communities — matching people for debates and discussions.
  • Ran heavy jobs in parallel with a hybrid thread/process pool, and exposed everything through FastAPI: users, content, messaging, recordings, automatic recaps.
  • Knowledge Graphs
  • RAG
  • SentenceTransformers
  • Hierarchical Clustering
  • LLMs
  • Web Scraping
  • FastAPI
  • MongoDB
  • MinIO
  • Python
node · text-to-video 2023 — now

Epineon.AI · Casablanca · Data Scientist · 2023—now

Turning a block of text into a finished video, automatically

A text-to-video platform: sentence segmentation and entity recognition pick the moments worth showing, then matching images and clips are assembled into a cut — no editor required.

  • Built the full pipeline end to end — collection, cleaning, pre-processing, S3 storage, and rendering.
  • Wired recognition over both images and video to choose visuals that fit the script.
  • NLTK
  • spaCy
  • MoviePy
  • Flask
  • FastAPI
  • TypeScript
  • AWS
  • Ollama
  • vLLM
  • SearXNG
node · misinformation 2022 — 23

Aalto University · Espoo, Finland · Data Scientist · 2022—23

Mapping how COVID-19 misinformation spread across Finland

My engineering final-year project: a data-driven study of COVID-19 misinformation and vaccination hesitancy — classifying the claims, then tracing the network that carried them.

  • Classified misinformation and vaccination-hesitancy content across Finnish social media with FinBERT.
  • Detected malicious bot accounts, then used social network analysis and community detection to surface the groups driving the spread — and visualised who amplified what.
  • FinBERT
  • Botometer
  • Social Network Analysis
  • NetworkX
  • Leidenalg
  • Gephi
node · chatbot 2022

CDG Prévoyance · Rabat · Backend Developer & Data Scientist (intern) · 2022

The NLP model behind a customer chatbot

Built the language model powering a support assistant — from understanding and cleaning the data to text analysis and intent prediction, shipped as a Streamlit app.

  • Python
  • NLP
  • Streamlit
  • PL/SQL

Cluster · 02 — Stack

The tools, by community

Grouped the way the graph groups them — five clusters that show up across the work.

ML / Deep Learning

  • PyTorch
  • TensorFlow
  • scikit-learn
  • Hugging Face Transformers
  • CNN

NLP / LLMs / RAG

  • RAG
  • spaCy
  • NLTK
  • FinBERT
  • SentenceTransformers
  • Ollama
  • vLLM

Graphs / Networks

  • Knowledge Graphs
  • Social Network Analysis
  • NetworkX
  • Leidenalg
  • Gephi
  • Clustering

Engineering / APIs

  • Python
  • Django
  • FastAPI
  • Flask
  • Java
  • TypeScript
  • R

Data / Cloud

  • Web Scraping
  • PostgreSQL
  • PL/SQL
  • MongoDB
  • MinIO
  • AWS
  • GCP
  • Spark
  • Kubernetes
  • Talend

Viz / Delivery

  • Power BI
  • Gephi
  • MS Project
  • Trello

Cluster · 03 — Background

Education & credentials

  1. 2018 — 23

    Master's degree — Computer Engineering

    Université Internationale de Rabat — Salé, Morocco

  2. 2022

    Exchange Program — Computer Science

    Aalto University — Espoo, Finland

Certificates

  • Advanced Learning Algorithms — Stanford University · Coursera · 2025
  • Supervised Machine Learning: Regression & Classification — DeepLearning.AI · Coursera · 2025

Languages

  • Arabic — mother tongue
  • French — bilingual
  • English — advanced

Edge · 04 — Connect

Let's build something
worth mapping.