Founder and a Product Manager with 9+ years of experience building innovative fintech and AI/ML solutions. Currently implementing AI/ML and Growth/KYC solutions for customers at a fintech startup. Previously at JP Morgan, scaling products for millions of users.
My background spans from software engineering to product leadership, with deep expertise in entrepreneurship, AI/ML integration, fintech, and data-driven product development. I've successfully launched and scaled products that serve millions of users. I've also built and scaled free dating app achieving 400% user base growth in 2 weeks with zero market spend.
Deployed Gemini 2.0 ML models for authenticity/confidence scores in legacy documents, generating confidence scores to preemptively identify and mitigate fraud risks. Saved 90k a month by implementing an in house confidence score solution and getting rid of an over priced vendor. Utilized AWS Rekognition to detect age and flagged the user if submitted date was off by a 3 year distance.
Built and scaled free dating app from concept to 400% user growth, implementing data-driven growth strategies, user engagement features, and a strong marketing strategy.
Predicting whether or not a person survived the Titanic disaster, based on their features.
This is a simple machine learning project that uses a decision tree to predict the survival of a person on the Titanic.
Type: Machine Learning
Tools: Python, Jupyter
Source: github.com/omaranwarny/titanic_data
Visualizing police data across the United States. This project uses Tableau to create a map of the United States and visualize the data.
Type: Data Visual
Tools: Tableau
Source: ucr.fbi.gov
Analyzing county-level economic data using OpenRefine and Tableau. This project uses OpenRefine to clean and transform the data, and then uses Tableau to create visualizations.
Type: Data Analysis
Tools: OpenRefine, Tableau
Source: https://www.kaggle.com/datasets
Analyzing historical weather data in Central Park, New York City. This project uses Python and Bokeh to create an interactive visualization of the weather data.
Type: Data Analysis
Tools: Python, Bokeh, PDFtables, Excel
Source: www.weather.gov
Real-world lessons from deploying ML models at scale, based on implementations at JP Morgan and a fintech startup.
Example of pygal World map. Source code available on https://github.com/omaranwarny/moonman
An interactive visual with a box plot. This box plot checks distribution of the dataset and detects any ouliers.
This is a text file character frequency counter created with Python. The program goes through a text file and counts every character and outputs a csv file. One of the many reasons why I prefer Python over other languages is it's rich libraray. Python is all about less coding and more work. The output csv can be imported in to Excel for a quick visual analysis.
A pygal samples that shows my expenses in percentage for the year 2015.
OpenRefine previously known as GoogleRefine is a great tool for cleaning and advance tranformation of datasets. The data can be transformed into different formats such as tsv, csv, html and JSON. The application is free and opensourse. One of the main reasons I use OpenRefine is beacuse of recipes. Recipes hold data structure info in order to be applied on future data tranformation. This saves time when working on data ETL'S
Tableau is a strong data visualiztion tool used mostly for Business Intelligence Here is an example of how to assign geo location to a specific type of data.
Bokeh is a python library for interactive data visualization on web pages. Python does not have many strong interactive visualization tools. Bokeh can be compared to tools such d3.js.
Bokeh can easily create JSON objects and output ready HTML files for the browser. Bokeh is native to python and can stream line live data. This is an example of how to create a line chart.
Bokeh Library:docs.bokeh.org/en
Excel is a strong tool for doing data analysis especially financial data. The user interface makes it easier to understand data and perform calculations. The complete workbook for this project is available on google drive.