Faran Ahmed Yasir - Machine Learning Engineer

Developed Vehicle detection in multiple cameras for theft protection and detection bypassing. Completed person dressing model to impose charges if dressing code is not followed. Collected big data of 12000+ images including Emirati, and other clothing using multiple sources. Collaborated with teams to create an attendance system using AWS Rekognition service, designed. Deployed models on AWS integrating software solutions to keep SWE fundamentals.

Developed Lead Generation and a recommendation system to find the best candidates. Classified resumes to recommend against given job description alleviating consultants. Achieved accuracy of 77 percent over Linear SVM and visualized results in Linear SVC classifier. Used profiles to better classify and recommend the best available training courses to the registered users. Email marketing using lead generation techniques deployed in the system to improve business.

• Supervised/Un-Supervised model development, testing & validation for multiple use cases • Predicted and optimized strategies for social media text & image processing • Created charts in Jupyter Notebook to perform preliminary analysis and visualize data using Matplotlib • Originated & Implemented a sentiment analysis tool to recommend an ideal cluster social profiles for marketing services

• Implemented regression analysis for Academic Social Network data • Analyzed citations developing same ideas and replicate their work • Designed PowerBI dashboards for analysis reporting • Developed and managed APIs using Python Flask and deploy on AWS • Designed cross-platform front ends using HTML/CSS and JavaScript (Angular)

Movie Business Predictor - 2017 • Scraped and labeled/preprocessed movie data from IMDB, BoxOfficeMojo. • Random Forest classification model trained with statistically labeled data. • Prediction for real-time movies with an accuracy of 76% achieved. • Added some ratio of a movie sentiment analysis on Social Media (Twitter and YouTube) trailer. • Recommended actors and actresses based on their growth in a genre scraping from BoxOfficeMojo.