Andraws Rasho

I am dedicated and passionate about technology and actively seeking opportunities in the field of technology. I bring a talent for problem-solving and a strong focus on customer service. With a particular interest in areas such as Cybersecurity, networking, software engineering and artificial intelligence. My goal is to leverage excellent collaborative abilities as a dedicated team player in any given project. I am eager to contribute my skills and enthusiasm to make a meaningful impact through diligent work.

Project 1: Relational Database Development with PostgreSQL
Description:
● Developed a relational database system for the Chicago Police Department to manage
data regarding a gang of bank robbers in Cook County.
● Designed and implemented the database schema using PostgreSQL, ensuring optimal
data normalization and integrity constraints.
● Created tables for various entities such as Banks, Robbers, Skills, HasSkills, HasAccounts,
Robberies, Accomplices, and Plans.

Project 2: Library System Transaction Processing with JDBC
Description:
● Developed a library management system focusing on database transaction processing
and locking mechanisms.
● Implemented a Java program using JDBC to interact with a PostgreSQL database, enabling
the management of books, authors, and library customers.
● Ensured data consistency and integrity by allowing multiple instances of the program to
access and update a common database concurrently without inconsistencies.

Project 3: Image classification using CNN model
Description:
● Built a neural network (CNN) to classify images of tomatoes, cherries, and strawberries
for an image classification task, using a dataset with 6,000 images.
● Conducted exploratory data analysis (EDA) and applied image preprocessing techniques,
including resizing, normalization, and data augmentation, to improve data quality and
enhance model accuracy.
● Developed and trained a baseline multilayer perceptron (MLP) as well as a custom CNN
model using PyTorch, comparing the performance of both models. Fine-tuned the CNN
through hyper-parameter optimization and experimented with various loss functions,
activation functions, and optimization techniques.

• Nyarie Muza, Cityfitness, +64 21 265 7102
• Clair, Capability and Development, +64 27 544 5203
• Thair, Case Manager, +64 21 962 809