Applied ML Engineer & Data Scientist

Imed Eddine Aouidane.

I turn clinical and business data into ML systems people can use.

I'm a master's-trained data scientist building computer vision, forecasting, and automation projects with clear evaluation, practical interfaces, and enough engineering discipline to move beyond a notebook.

Health AI Computer vision Forecasting Data products
112k+ medical images used in X-ray modeling
<200ms surveillance inference latency target
WMF 2025 AI innovation conference, Bologna
View Selected Work Resume
Available for applied ML roles, research collaborations, and selected freelance projects.
Portrait of Imed Eddine Aouidane
WMF 2025 AI conference, Bologna
Clinical AI Hospital workflow project

About Me

I'm an ML Engineer and Data Scientist from Algeria with a Master's in Data Science for Economics and Business. I like the parts of machine learning where research discipline meets shipping: clean data, honest evaluation, useful interfaces, and a clear reason for the model to exist.

My work spans healthcare, oil and gas, e-commerce, and applied AI. Recent projects include a cardiology assistant for a hospital workflow, chest X-ray classification pipelines, climate forecasting experiments, and client-facing analytics applications.

I'm especially interested in health AI and applied research that helps people make better decisions under uncertainty.

  • PyTorch, TensorFlow & scikit-learn
  • Faster R-CNN, YOLO & ResNet
  • ARDL, ECM, ARIMA & forecasting
  • FastAPI, Streamlit & Telegram bots
  • PostgreSQL, Notion API & data pipelines
  • Python, R, Docker & Git workflows

Skills & Expertise

A practical stack for applied ML work: models, evaluation, data plumbing, and the communication needed to turn results into decisions.

Applied ML

  • Deep learning pipelines
  • Model evaluation and diagnostics
  • Feature engineering
  • Production-minded experimentation

Computer Vision

  • Faster R-CNN, YOLO, ResNet
  • Medical image classification
  • Object detection and surveillance
  • Grad-CAM and explainability

Forecasting

  • ARDL and ECM frameworks
  • ARIMA, SARIMA and ETS
  • XGBoost and LSTM comparisons
  • Stability and residual diagnostics

Data Systems

  • PostgreSQL and normalized schemas
  • Web scraping and ETL
  • FastAPI and Streamlit apps
  • Notion and Telegram integrations

Delivery

  • Client discovery and scoping
  • Clear technical communication
  • Independent project lifecycle
  • Documentation and iteration
Research and recognition

Evidence first. Demos second.

My strongest projects pair statistical discipline with practical engineering: causal modeling for AI revenue impact, clinical decision support, medical computer vision, and forecasting for climate risk signals.

01

Ministry-recognized founder

Auralytics AI was recognized as an innovative project by Algeria's Ministry of Knowledge Economy and Startups.

02

Master's thesis with econometrics

Used ARDL and ECM methods to model the link between AI adoption and NVIDIA financial performance.

03

International AI exposure

Presented and networked at We Make Future 2025 in Bologna with researchers and industry teams.

Selected Systems & Research

A curated set of projects that show range across research, deployed systems, client-facing analytics, and product thinking.

Health AI

Cardiopulmonary Classification from Chest X-rays

Python / PyTorch / ResNet-50 / OpenCV / Grad-CAM

Owned End-to-end modeling Signal 112k+ image dataset

Built a multi-label deep learning pipeline for cardiomegaly, pneumonia, and pleural effusion detection using the NIH Chest X-ray dataset with 112,000+ images and 14 disease labels.

AUC-ROC and precision-recall evaluation Grad-CAM model attention visualizations
Clinical Assistant

CardioBot - AI Cardiology Assistant

Python / Telegram Bot / Claude API / Notion API

Owned Assistant workflow Signal Clinical record retrieval

Designed a Telegram-based assistant for a cardiologist at CHU Mustapha Hospital, enabling natural-language access to structured patient records, medication notes, and clinical summaries.

Designed for active hospital workflow Real-time retrieval from structured records
NVIDIA revenue forecasting study
Master's Thesis

NVIDIA AI Revenue Impact Study

R / ARDL / ECM / ARIMA / Econometrics

Owned Econometric thesis Signal ARDL/ECM validation

Modeled the causal relationship between AI adoption and NVIDIA revenue using quarterly financial data from 2015-2024, with ARDL bounds testing and error-correction modeling.

F-statistic 8.47, p < 0.001 ~67% quarterly adjustment speed
Climate ML

Climate Forecasting & Extreme Weather Detection

Python / SARIMA / XGBoost / LSTM / pandas

Owned Forecasting comparison Signal Extreme-event flags

Compared statistical and ML forecasting methods on daily climate data to predict temperature trends and flag extreme heat events using engineered lag and interaction features.

SARIMA, XGBoost and LSTM comparison Heatwave anomaly detection module
Real-time surveillance computer vision
Computer Vision

Real-Time Multi-Camera Surveillance System

PyTorch / TorchVision / PostgreSQL / Flask

Owned CV inference stack Signal PostgreSQL event logging

Built a GPU-optimized Faster R-CNN inference pipeline with multi-threaded camera management, PostgreSQL event logging, and automated alert generation.

Sub-200ms end-to-end latency target Production-style monitoring architecture
E-commerce intelligence dashboard
Client Data Product

E-commerce Competitor Intelligence

Python / PostgreSQL / Selenium / Power BI

Owned Scraping and BI pipeline Signal 2,500+ records

Scraped and normalized 2,500+ product records from a cosmetics e-commerce competitor, enabling structured market positioning and competitor benchmarking.

Live Demo
Oil and gas predictive analytics
Predictive Analytics

Oil & Gas Predictive Analytics Platform

Python / Streamlit / scikit-learn / seaborn

Owned Streamlit analytics app Signal Export-ready reports

Developed a Streamlit application for bottom-hole pressure prediction with EDA, model performance reporting, and client-ready export workflows.

80% train/test accuracy reported Automated report generation
Auralytics AI platform
Founder Project

Auralytics AI Platform

OpenAI Whisper / NLP / Python / FastAPI

Owned Founder product build Signal Ministry-recognized

Founded and built an AI meeting platform with real-time transcription, speaker diarization, summarization, secure authentication, and data workflows.

Ministry Recognized Brand Case
Medy Wear brand identity
Creative Venture

Medy Wear Brand Identity

Brand Strategy / Logo Design / Packaging / Apparel

Owned Brand direction Signal Creative range

A separate creative venture showing visual direction, brand building, and product taste. It supports the portfolio without competing with the main ML narrative.

View Gallery

Experience, Education & Recognition

2024 - Present

Freelance Data Scientist

Scope client problems, build predictive and automation systems, explain findings clearly, and manage projects from briefing through delivery.

Jun 2025

Master's Degree - Data Science for Economics & Business

Graduate School of Economics, Oran. Thesis focused on ARDL and ECM econometric frameworks for modeling AI adoption and financial performance.

Jun 2025

We Make Future 2025 - Bologna, Italy

Represented Algeria's Graduate School of Economics at an international AI and innovation conference, engaging with researchers and industry teams.

Feb - Apr 2025

Data Analyst Intern @ MDM Hasnaoui

Prepared industrial data for predictive use cases, connected production data with KPIs, and diagnosed data quality issues before modeling.

Dec 2023

Data Analyst Intern @ LITMED PRO

Analyzed and visualized patient-level healthcare datasets, supported process optimization, and cleaned structured medical records for exploratory analysis.

Get In Touch

I'm available for applied ML roles, research collaborations, and selected freelance work in health, climate, forecasting, and data systems.

aouidaneimad@gmail.com +213 559 044 133 LinkedIn Download CV