ResuGrade website
Website Development front-end and NLP app back-end built using LangChain and OpenAI GPT3.5 and GPT4 model APIs.
Website Development front-end and NLP app back-end built using LangChain and OpenAI GPT3.5 and GPT4 model APIs.
Simple dashboard and tool designed for real-time data analysis and pattern recognition for crypto data. It takes the last hour of crypto data, generates vector embeddings, and compares them against historical embeddings stored in a Pinecone vector database using cosine…
An automated data extraction and collation tool merging data from 2 publicly available sources to generate a complete data set for each individual property needed. A specific automated web scraping process was required due to API unavailability.
ACT Lab is a safety and product testing lab with offices all around the world. After each safety test, the results report must be submitted to their clients and vendors. One of their vendors is a global retailer with very…
Web application which allows an user to track and display positive or negative Twitter sentiment for a keyword or hashtag. WordPress was used for the front end and Python & TensorFlow were used for neural network training and prediction. Project…
Automated data collection and transformation of data from CBDCTracker.org. Outputs an Excel spreadsheet of the data and a report focused on changes and trends.
Custom web application with Python back end which converts NOAA ocean data files (.grib) containing wave height, wave period, wind direction/intensity, and water temperature, into numerous highly accurate visual representations. Automated download, process, and upload few gigabites of data every…
Data collection and processing application built in Python to aid in the due diligence process of a real estate investment company. It automatically gathers and processes over 30 data points from a variety of sources to create a complete and…
A machine learning algorithm trained on over 215,000 loan data records, which predicts the probability of default at a higher accuracy than the existing model. A series of algorithms like Random Forest, Logistic Regression, XGBoost, Gradient Boost, and Ensembles, were…
Australia fires visualization project used as social media awareness in an effort to raise donations for the fire victims.
A 3d rendering of merging a high-resolution land GIS dataset with a bathymetry dataset of a lower resolution.