Arooj.
Kanvis
Arooj Fatima

Arooj Fatima

CS Student at LUMS | Specialized in High-Performance Data Structures & ML Frameworks

Hiring? Every applicant arrives already interviewed.

Overview

Computer Science student at LUMS with a strong foundation in complex data structure implementation and machine learning. Proven ability to translate abstract logic into efficient C++ and Python code, with a focus on structural visualization and algorithmic precision. Currently seeking to apply technical problem-solving skills to the startup ecosystem at NIC Lahore.
Lahore, Pakistan

Project Portfolio

PERSONAL

Present

Machine Learning Implementation Framework

Student Developer

Implemented and evaluated core ML algorithms spanning probabilistic classifiers, regression models, and neural networks. Worked through the full design-implement-evaluate cycle: loss functions, gradient descent optimization, hyperparameter tuning, and k-fold cross-validation. Derived and applied SVM Lagrangian formulations, information-gain-based decision tree construction (ID3), and backpropagation for neural networks.
PythonMachine LearningRegression modelsNeural networksSVMDecision treesID3Backpropagation
PERSONAL

Present

C++ High-Performance Data Structures Library

Student Developer

Implemented a doubly linked list, node- and array-based stacks and queues, and applied them to build Warehouse and RadarSystem simulation modules. Built a binary search tree and a self-balancing AVL tree from scratch, implementing recursive insertion, deletion, search, height tracking, and all four rebalancing rotations (LL, RR, LR, RL), and integrated them into a Catalogue application. Utilized manual structural visualization and diagramming to map out complex pointer reattachments and subtree rotations. Implemented hash maps with multiple collision-resolution strategies (chaining, linear and quadratic probing, double hashing) and graph data structures with traversal algorithms (BFS, DFS) for solving graph-based problems. Analyzed time and space complexity trade-offs across implementations and debugged pointer-level issues such as subtree reattachment during rotations and reference-to-pointer deletion patterns.
C++Doubly linked listStacksQueuesBinary search treeAVL treeHash mapsGraphsBFSDFS
PERSONAL

Present

Distributed Systems & Network Protocol Suite

Student Developer

Developed client-server socket-based network applications, handling message marshaling and packet-level communication between hosts. Designed and implemented a reliable data transfer protocol over an unreliable channel, addressing packet loss, acknowledgments, and retransmission strategies. Built distributed application components covering load distribution, coordination, replication, and failure handling across a multi-stage development sequence.
PythonSocket programmingClient-server architectureReliable data transfer protocolDistributed systems

Education

Freshman Year (Transfer)

Beaconhouse National University

Relevant Coursework

Introduction to Programming (CS 200)Calculus I & IIDigital Logic Circuits

2028

BS Computer Science

Lahore University of Management Sciences (LUMS)Specialization in Computer Science

Relevant Coursework

Data Structures (CS 202)Network Centric Computing (CS 382)Foundations of AI & Machine Learning (CS 331)Introduction to Programming (CS 200)Calculus I & IIDigital Logic Circuits

Skills & Interests

Algorithms

Data Structures

C++

Stacks

Queues

Binary search trees

AVL trees

Hash tables

Socket programming

Heaps

Priority queues

Python

Visual Studio Code

Graphs

Slack

Dijkstra's algorithm

Sorting algorithms

Big-O analysis

Linked lists

LMS collaboration workflows

Linear regression

Logistic regression

B+ trees

Tries

Gradient descent

Regularization

Naive Bayes

k-Nearest Neighbors

Support Vector Machines

Decision trees

Neural networks

Networking

Distributed Systems

HTTP

DNS

TCP

Git

Machine Learning

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