Final-year engineering student · Rabat, Morocco

Saad Charifi

AI & Data EngineeringandQuantitative Finance

Building intelligent systems from models to production, with a focus on AI engineering and quantitative finance.

01 — About

An engineer who builds the whole system — data, models, retrieval, orchestration and interface.

I'm a final-year Computer Science engineering student at ISGA Rabat, specializing in Big Data. My work sits where data engineering meets machine learning: I take systems from raw data to models to the interfaces people actually use.

Financial markets are the domain I keep returning to — noisy, non-stationary and unforgiving, which makes them a demanding testbed for applied AI. My long-term direction is to combine AI engineering with quantitative finance and market research.

01Computer Science · Big DataEngineering foundation
02Machine & Deep LearningPredictive modeling
03Generative AI · RAGRetrieval & LLM systems
04Agentic AIOrchestration & automation
05Quantitative FinanceApplication domain

02 — Selected work

Complete systems, not isolated notebooks.

Three projects across predictive ML, retrieval-augmented generation and deep-learning NLP — each built end to end.

01FlagshipPersonal project · Source private

C7 — Multi-Model AI System for Financial Market Analysis

Specialized models, explicit agreement, explained output.

An end-to-end AI ecosystem that runs several specialized machine-learning models on financial and macroeconomic data, measures where their signals converge or diverge, and passes the result through an LLM supervisory and synthesis layer that produces a risk-aware, explained analysis delivered through a web dashboard.

Predictive AI
Agentic orchestration
Generative AI
Financial & macro data
Infrastructure & web delivery

Specialized models

Relative return prediction
Macroeconomic regime classification
Market sentiment / risk regime classification
Volatility / risk modeling

Stack: Python · Pandas · NumPy · scikit-learn · XGBoost · LightGBM · Random Forest · n8n · JavaScript · JSON · PHP / MySQL · Linux / Ubuntu · VPS

View case study Architecture Screenshots DemoSource code private

System architecture

User
Web interface
Backend
n8n orchestration
Financial / macro data
Specialized ML models
- Relative return
- Macro regime
- Sentiment / risk regime
- Volatility / risk
Individual signals
Convergence / divergence analysis
LLM supervisory & synthesis layer
Backend
Dashboard + final analysis

02Internship · 2 months

Enterprise RAG Platform — MEDZ

Two-stage retrieval behind a production web platform.

A web platform developed during a two-month internship: a client-facing interface, a PostgreSQL data layer, geographic visualization, and two AI assistants — technical and navigation — backed by a retrieval-augmented generation pipeline with bi-encoder retrieval, cross-encoder reranking and Gemma 2.

Pipeline

User query
Embedding
PostgreSQL + pgvector
Top 25 candidates
Cross-encoder reranking
Top 5 contexts
Gemma 2
Generated answer

Stack: Next.js · React · PostgreSQL · pgvector · Prisma · LangChain · BGE-M3 · Cross-encoder reranking · Gemma 2 · Ollama

Case study Architecture Screenshots GitHub

03Personal project

Financial Market Sentiment Classification

Classifying the dominant sentiment of financial news.

A deep-learning NLP system that classifies the dominant sentiment of financial news. Custom embedding implementation with NumPy and neural-network training using PyTorch/CUDA, with MLP and self-attention components.

Pipeline

Financial news text
Tokenization
Embeddings
Self-attention
MLP · hidden layers
Softmax classification
Dominant sentiment

Stack: Python · NumPy · PyTorch · CUDA

Case study ArchitectureGitHubSoon

03 — Experience

Where the work happened.

2026/07/01 — 2026/08/31
Internship
AI & Full-Stack Development Intern · MEDZ
Individual implementation under professional supervision.
- Redesigned and developed the company web platform with a professional, client-facing interface.
- Built the PostgreSQL data layer with Prisma ORM, plus mapping / geographic visualization.
- Implemented a technical assistant and a navigation assistant on a RAG pipeline: BGE-M3 embeddings, pgvector retrieval, cross-encoder reranking and Gemma 2 served with Ollama.
2025 — Present
Independent initiative
Founder — Quantitative Research & AI · Wolfbank Partners
A startup based in Kenitra, Morocco, applying AI, machine learning and algorithmic trading to financial markets.
- Trading and development of automated trading systems.
- AI research: applying machine-learning models to market and macroeconomic data, with an emphasis on risk analysis.
- Long-term investment strategies.
- Developing the systems and tooling that support this research.

04 — Technical skills

Tools I build with.

Grouped by where they sit in a system. Methods and concepts are documented inside each project case study.

Machine Learning & Deep Learning

Python
scikit-learn
XGBoost
LightGBM
PyTorch
CUDA

LLM, RAG & Agentic AI

LangChain
LangGraph
LangSmith
n8n
Ollama
RAG pipelines
Retrieval & reranking

Data

Pandas
NumPy
PostgreSQL
pgvector
Prisma

Web & Backend

Next.js
React
JavaScript
PHP

Infrastructure

Linux / Ubuntu
VPS
Git / GitHub

05 — Education

Engineering foundation.

ISGA Rabat

Software Engineering AI & big Data

Specialization: AI & Big Data

Status: Final year
Graduation: Expected graduation 2027
Location: Rabat, Morocco

06 — Certifications

Continuing education.

Trading Using Technical Analysis
Issuer
CFI
Status
Active
Machine Learning & Deep Learning Applied to Finance
Issuer
CFI
Status
Active
Risk Management / Monte Carlo Simulation
Issuer
CFI
Status
Active
CFA level 1
Issuer
CFA Institute
Status
Candidate
Credential
URL

07 — Contact

Let's build intelligent systems.

Open to PFE internship opportunities in AI / ML engineering and AI applied to financial markets.