Ahmed Magdy
[email protected] • Medellin, Colombia
Computer Science Engineer, love everything related to technology, interested in Machine learning/AI, also a bit of gamer.
Education
Helwan University, Bachelor of Engineering Sep 2018
Major: computer science
GPA: very good (the tenth)
Work Experience
Machine Learning Engineer at AI Fund, Colombia, Oct 2019 - present
- Rustle.co
- Developed a customer service AI platform for better customer support quality.
- Developed an automated data pipeline for extracting/scraping data from CS websites using MongoDB and AWS with S3.
- Deployed a Question-Answering NLP system over HTML text to answer CS questions given a company knowledge base using AWS and Fast-API.
- Deployed a Search Engine system using Elasticsearch engine.
- Factored.AI
- Implemented a Zero-Shot classification system for various use cases using NLP SOTA, dockers and Fast-API.
- Worked on creating a Curriculum Vitae(CV) parsing system where it can extract all the needed information from it using NER modules and other NLP algorithms.
- Created an HR Matching tool , helps to match candidates with job descriptions and gives a score to each candidate with a proper explanation to each score.
AI Engineer at Integrated Development, Egypt, Nov 2018 - Apr 2019
- Developed Arabic NLP modules like chatbots, intent detection and NER.
- Helped in developing an online travel agency platform using AI technology to facilitate and improve user experience, solving problems like record linkage database with ML.
- R&D in stock market prediction with reinforcement learning and time series forecasting.
- Developed an automated testing system for stock market predictions using trading indicators and hypothesis testing.
Machine Learning Developer at Egyx.ai, Egypt, May 2017 - Aug 2018
- Developed Arabic Chatbots for restaurants customer services.
- R&D in AI Arabic home assistant.
Projects
Novel Ensample Learning Approach to Unsupervised Record Linkage
- Implementing a semi supervised approach to solve record linkage database problem.
- The proposed approach is to train a self learning adding automatic seed selection.
- Achieving around 85% accuracy outperforming simpler methods like TF-IDF.
Tweets Author verification
- Research project in which I try to outperform some of the benchmarks with simpler methods.
- Calculating Z scores for top N-words and compare them together.
- Used SVM classier with a different set of features like lexical diversity and number of hashtags.
- Used one-shot learning with the Siamese network to achieve 87% accuracy.
MASRI(GP)
- A home assistant that takes voice commands and perform services.
- Controlling smart devices , playing music and scheduling tasks etc.
- Responsible for Voice recognition modules, trigger word, Text To Speech.
- Deployed the services on Heroku Server with Django.
- The project won IEEE Hackathon and joined the Nile University accelerator program and got funded.
Mwaslaty App
- Android App helps the user to find the best transportation to reach a certain destination.
- worked with PHP to deal with the database using SQL to insert and acquire data.
- Deployed the service on backend server with PHP and Django.
Awards and Certificates
- A graduate of Next Technology Leader(NTL) program which is held every year in Smart Village Egypt.
- Over 20 Coursera and Udacity courses on machine learning, python programming, data analysis
Technical Skills
Programming
python, C/C++, R, SQL, mongoDB
Machine Learning
linear/logistic regression, KNN, SVM, randomforest, XGBoost, sklearn
Deep Learning
RNN, CNN, transformers, keras, pyTorch, tensorFlow, huggingface
Data Analysis
statistical tests, P-value, visualizations, descriptive/inferential statistics, plotly, matplotlib, scipy
Mathematics/Statistics
vectors, matrices, differential/integral calculus, random variables, probability distributions, Bayesian statistics
Miscellaneous
Linux/Ubuntu, AWS stack, docker, fast-API, scrapy, Elasticsearch
