Salman.
Kanvis
Salman Abid

Salman Abid

Software Engineering Student at COMSATS | AI & ML Enthusiast
Hire Me

Overview

Software Engineering student at COMSATS with a focus on Deep Learning and Computer Vision. Developed a Leukemia classifier with a 99% F1-score and architected full-stack management systems. Driven by solving complex technical challenges and optimizing performance in data-intensive environments.
Lahore, Pakistan

Media Showcase

Project Portfolio

ACADEMIC

Present

Formula 1 Team Management Simulator

Developer

Developed an F1 management simulator where users recruit drivers within a budget and manage a full racing season. Implemented probability-based race outcome simulations using driver statistics, skill ratings, and performance metrics. Applied AVL Trees, Tries, and sorting algorithms for optimized data storage, retrieval, and leaderboard management with file-based persistence.
C++File HandlingAVL TreesTriesSorting Algorithms
ACADEMIC

Present

Financial Advisor System

Developer

Developed a Java-based financial management system using OOP principles for tracking income, expenses, savings, and financial goals. Integrated SQL database for secure storage and retrieval of user financial records, with HTML report generation for selected time periods. Implemented dual-module architecture with a user dashboard and admin panel for activity monitoring and account management.
JavaSQLHTML/CSSJS
PERSONAL

Present

Honda Dealership Management System

Lead Developer

Developed a full-stack dealership management system covering vehicle inventory, sales, financing, spare parts, and service operations. Designed a normalized 10-table SQL Server database with stored procedures, triggers, indexes, and views for transaction integrity. Integrated Python with SQL Server via pyodbc for VIN-level vehicle tracking, real-time inventory sync, and dealership reporting.
PythonMicrosoft SQL Serverpyodbc
Code
PERSONAL

Present

AI Leukemia Classifier

Collaborator

Collaborated to classify blood cells into 7 categories (ALL leukemia + 6 normal types) across 3 merged datasets, achieving 99% macro F1-score. Built EfficientNetB3 with BatchNorm, Dense(256, L1/L2), Dropout(0.45), and Softmax; trained with Adamax and stratified 80/12/8 split. Resolved GPU memory crashes (12× speedup), cross-dataset inconsistencies, and checkpoint failures across 4 training iterations.
PythonTensorFlowKerasKaggleEfficientNetB3Adamax
Code

Education

2024 – Jun 2028

Bachelor of Science in Software Engineering

COMSATS University Islamabad, Lahore CampusSpecialization in Software Engineering
GPA: 3.61 / 4

Relevant Coursework

Data Structures and AlgorithmsObject-Oriented ProgrammingDatabase SystemsArtificial Intelligence

Impact & Recognition

Professional Certifications

Harvard University

CS50 Introduction to Programming

Credential f9f7aa83
Harvard University

CS50 AI

Credential d72bd427

Skills & Interests

Python

Machine Learning

SQL

Google Colab

C++

TensorFlow

Keras

Microsoft SQL Server

Git

GitHub

VS Code

Java

Deep Learning

Computer Vision

Transfer Learning

Model Training & Evaluation

JavaScript

HTML/CSS

IntelliJ IDEA

SQLite

MySQL

Kaggle

File Handling

Let's Connect

I'd love to hear from you. Feel free to reach out.

These unlock once you finish onboarding and publish your profile.

More people on Kanvis

Browse more people