Hi, I’m Rachid,

I’m a Software Engineer with over 5 years of experience in Backend development with Python, with a strong interest in image processing and object detection.

My work focuses on creating systems that reduce inefficiency, save time, and solve real-world problems.

My Recent Work

Here are a few past development projects I’ve worked on.

#Aeronautics

Project 1 : Aircraft components quality defects detection

Problem:

Rising client complaints about the quality of aircraft components produced were jeopardizing trust and operational efficiency. Identifying defects manually was time-consuming and prone to errors, leading to delays and dissatisfied customers.

Challenge:

The production line involved inspecting a huge number of components with intricate quality requirements, making it difficult to accurately distinguish defective parts from compliant ones.

Solution:

Defect Detection Software: Developed a robust software system to detect major defects in airplane components using advanced image processing algorithms.

Calibration Processes: Designed intuitive UIs and streamlined workflows to calibrate inspection stations for optimal performance.

Production Deployment: Deployed the system directly at the production site to ensure seamless quality control at the source.

Training Program: Trained the company operational teams to effectively use the application and integrate it into their daily processes.

Impact:

Client complaints decreased significantly due to the early detection of defects before shipment, and inspection speed improved by 97%, enabling faster production cycles and enhanced customer satisfaction.

---

#Transport

Project 2 : Rental car damage inspections

Problem:

Car rental agencies rely on manual car damage inspections before and after rentals, using paper-based contracts to mark damage locations. This approach is time-consuming, non-specific, and prone to errors, leading to inefficiencies and potential disputes with customers.

Challenge:

Collecting car damage data and identifying the best machine learning model to detect and assess car defects accurately required extensive testing and comparison across multiple models, including Detectron2, AutoSAM, and VGG19. After rigorous evaluation, VGG19 was chosen for its superior performance.

Solution:

Multi-Agency Platform: Designed and implemented a comprehensive car rental platform accommodating multiple agencies.

Automated Documentation: Established PDF contract generation for a seamless and standardized rental process.

Defected Car Detection: Introduced machine learning-powered car damage detection using VGG19 for accurate and efficient inspection.

Time Optimization: Reduced damage inspection time by 93%, enabling agencies to serve more customers with greater accuracy.

Impact:

Happy Cars transformed the car rental process, streamlining damage detection and documentation for agencies while delivering a faster, more reliable experience for their customers.

---

#Education

Project 3 : Students, From foggy and uncertain future, to clear and informed decisions

Problem:

High school students often face a foggy and uncertain future, struggling to understand the pathways that lead from their current academic stage to their desired careers. The lack of structured, accessible, and connected information leaves them overwhelmed and uninformed during critical decision-making moments.

Challenge:

Collecting and connecting all academic and professional programs in Morocco, spanning from high school diplomas to final career options, required tackling fragmented data sources across websites, PDFs, and unstructured content.

Solution:

Data Collection: Implemented web scraping, PDF parsing, and image data extraction to gather information from diverse sources.

Data Structuring: Leveraged AI models to structure and categorize data into a cohesive format.

Pathway Optimization: Utilized graph theory algorithms, including Dijkstra’s Algorithm, to visualize and suggest optimal routes from high school to career goals.

Accessibility Enhancement: Integrated a WhatsApp chatbot using Meta’s official WhatsApp API to provide students with limited internet access personalized recommendations and updates on educational programs.

Impact:

In less than one month, Sejlny attracted 20,000+ users, empowering students with structured guidance to make informed decisions about their academic and career journeys, regardless of their internet connectivity.