Amna Shamshad

Data Engineer | Analyst

Skilled in Python, SQL, Tableau, PySpark, Databricks and SAS @AmnaShamshad



A Computer Science graduate with Masters in Business Intelligence and Data Analytics - Fulbright Scholar

EXPERIENCE

Data Engineer Afiniti - Islamabad, Pakistan

As a Data Engineer at Afiniti, I automated the process of client revenue calculation by creating an ETL (Extract, Transform, Load) pipeline using Talend. This demonstrates my proficiency in data engineering and my ability to design and implement automated data workflows. I also performed data analysis to track optimization metrics by creating dashboards.

Associate Software Engineer - i2c Inc. Lahore, Pakistan

As an Associate Software Engineer at i2c Inc, I analyzed customer issues related to software products. This indicates my problem-solving skills and my ability to understand and troubleshoot software-related problems. I logged, prepared, and analyzed data using Informix to resolve customer issues. I was involved in data manipulation and data analysis tasks.

Data Science Intern - Higher Education Commission, Pakistan

As a Data Science Intern at HEC, I optimized data preparation code using multiprocessing, resulting in an 80% increase in performance. This demonstrates my skills in optimizing code for efficiency and my ability to leverage parallel processing to enhance performance. I also predicted traffic congestion for roads in Islamic and Rawalpindi using XGBoost with 96% accuracy













Luna Crash Analysis Using Social Network Analysis in R













In this project, I have used social network analysis in R in order to explore to what extent can Luna Crash be detected. I had Terra network data available on Standford data website that contains transactions for stable terra coins that include 6 stable coins USDT, USDC, DAI, UST, PAX and WLUNA. I explored the network properties for Luna network. The project includes

  • Graphical analysis of network for various months before and after Luna crash
  • Comparison of transitivity, reciprocity and centralization prior and post crash
  • Correlation between the six crypto coins
  • Checking for homophily effects in social network prior crash
  • Comparison of network modularity before and after crash
  • Comparison of price stats for various month


View Project

Tableau Projects

Dognition Company Data Analysis for Increased Revenue

In this project, I analyzed the Dognition company data to come up with a strategy that would increase the number of hours users spent on website for playing various games



These analysis suggested that Dognition should focus on the two tests after which there is a decline in the number of customers taking tests. More advertisements should be sent on Sunday and around 5-7pm throughout the week More advertisements should be sent on Sunday and around 5-7pm throughout the week. The company should not invest in free signups because there is no visible positive effect of free signups on number of tests completed by users

Salary Comparison of Various Data Related Roles Across US

In this project, I analyzed the pay disparity with respect to roles, states and companies in US.

Brain Tumor Classification Using Keras Python

This projects focuses on the classification of three types of brain tumor that include meningioma, glioma, pituitary tumor. I have used VGG-16 classifier and mobile nets for the classification of tumors using keras API. I have evaluated the performnace of the two classifiers using accuracy and confusion matrix

View Project

Credit Card Fraud Detection Using Python

I have created a program to correctly identify fraudulent transactions using credit card transactions data. The exploratory data analysis showed that the data was highly skewed. I applied Random Under Sampling Random Over Sampling and Smote Method to balance the data. I applied logistic regression, random forest, Gaussian Naive Bayes, KNN (K Nearest neighbour) and SVM (Support Vector Machine)

View Project

Traffic Congestion Prediction and Visualization in Python

This project includes the data collection process using on Open Source Routing Machine (OSRM) Server and the multiprocessing algorithm, used for improving the performance of code that processes data on OSRM Server. OSRM is a routing engine that provides (shortest) routes between origins and destinations on Open Street Map (OSM) based road networks. The Nearest Service by OSRM may be used to map GPS data on OSM road networks. The latitudes and longitudes (coordinates) from Floating Car Data (FCD) are sent to the OSRM server using nearest API to find the pair of nodes for the specific coordinates.The parallelization process yields pronounced results which have been discussed in my research paper

View Project

Get in touch