Software Engineer | Accountant | AI & Automation
I am a Software Engineer and CPA with 15+ years of experience in data science, software development, and financial planning & analysis. I specialize in leveraging AI, automation, and big data to optimize financial performance and business operations.
I am passionate about driving digital transformation by integrating finance and technology to create smarter, data-driven solutions.
124,000 nodes and 68 million connections
A user-friendly knowledge graph that simplifies drug safety research by mapping connections and calculating similarity scores between drugs, genes, targets, and adverse effects, enhancing efficiency and decision-making in pharmacology.
View Project
VR operating room for medical education using Oculus, allowing student to simulate intricate healthcare settings, explore the environment, interact with objects, and build confidence.
View Project
Generates a heatmap and CSV dataset of frequent transit stops using GTFS data, helping with resource allocation and service planning. Built with Scala and Spark for efficient data processing.
View Project
Utilized encoders to encode protein sequences and trained a model to calculate similarity scores between amino acid sequences. This model aids in identifying drug targets, enhancing drug design, and minimizing off-target effects.
View Project
Mathematical Group Theory demonstration apps in R Shiny Dashboard
View Project
A Python-based Django app for generating customer invoices with a built-in product and client database. Users can create profiles, add clients and products, select items from a dropdown when invoicing, generate PDFs for emailing, and mark invoices as paid.
View Project
This project implements a simple question answering system using natural language processing (NLP) techniques, specifically focusing on document and passage retrieval based on a user's query. The system uses TF-IDF (Term Frequency-Inverse Document Frequency) to rank documents and sentences in terms of their relevance to the query.
View Project
This project uses a Message Passing Neural Network (MPNN), a type of Graph Neural Network (GNN), to predict whether molecules like caffeine, sucrose, D-glucose, fructose, and acetaminophen can cross the blood-brain barrier.
View Project
This project implements a parser AI capable of parsing English sentences and extracting noun phrases using context-free grammar (CFG). The AI analyzes sentence structure and extracts noun phrase chunks, providing insights into the sentence's grammatical structure.
View Project
This project implements a neural network using TensorFlow to classify traffic signs from images. The model is trained on the German Traffic Sign Recognition Benchmark (GTSRB) dataset, which includes images of 43 different types of road signs. The AI is capable of identifying traffic signs in real-world scenarios, aiding in the development of autonomous driving systems.
View Project
This project implements an AI model to predict whether an online shopping customer will complete a purchase. The model uses a k-nearest neighbors (KNN) classifier to analyze user behavior based on various browsing metrics.
View Project
This project implements the PageRank algorithm to rank web pages by importance. The algorithm is inspired by the original Google PageRank method, which evaluates web pages based on their linking structure.
View Project
This project implements an AI that learns to play the game of Nim using reinforcement learning. The AI trains itself through thousands of self-play games, gradually improving its strategy by updating Q-values based on rewards.
View Project
This project is an AI-driven crossword puzzle generator that uses constraint satisfaction techniques to create valid crossword solutions based on a given puzzle structure and a word list.
View Project
This project implements an AI solver for the classic Minesweeper game using propositional logic and inference techniques.
View Project
This project implements a logic-based AI solver for "Knights and Knaves" puzzles, a type of logical puzzle introduced by Raymond Smullyan.
View Project
This project implements an AI for playing Tic-Tac-Toe optimally using the Minimax algorithm. The AI ensures that it never loses, making the game either a win for the AI or a tie if played optimally.
View Project
This program determines the number of "degrees of separation" between two actors based on the movies they have starred in. It finds the shortest path connecting two actors by identifying the sequence of movies that links them.
View Project
This app explores the isomorphism between two mathematical groups: SL2(F4) and A5. At first glance, these two groups might seem completely different, but they actually have the same structure! Here's a simple explanation of the connection between them.
Github View Project
This app explores the mathematical foundations of geospatial visualization, specifically focusing on great-circle navigation, vector transformations, and Mercator projections.
Github View Project
The Group D6 Enhanced app demonstrates the symmetry group of a regular hexagon, known as the Dihedral Group D6, through the permutations of the numbers 1 through 6. This app provides a powerful visualization tool for understanding group theory concepts, including subgroups, cosets, conjugation, and the generation of subgroups.
Github View Project
The Group D4 app demonstrates the Dihedral Group D4, which represents the symmetries of a square. This group consists of rotations and reflections that map the square onto itself. The app provides an interactive way to explore these symmetries and understand how they correspond to elements of the group.
Github View Project
This project implements an application that generates a random 3 × 3 matrix with entries from the finite field Z_5, computes its characteristic polynomial, determines its eigenvalues, and verifies the Cayley-Hamilton theorem. If the matrix has three distinct eigenvalues, the app also computes the corresponding eigenvectors.
Github View Project
This project is a simple web application that allows users to compute the product of two given permutations and display the powers of a permutation.
Github View Project
A real-time financial data pipeline using Kafka, Elasticsearch, and Kibana. The system enables year-over-year financial ratio evaluations for selected companies and visualizes the data using Kibana.
View Project
A real-time financial data pipeline using Kafka, Elasticsearch, and Kibana. The system enables year-over-year financial ratio evaluations for selected companies and visualizes the data using Kibana.
View Project
Generates a heatmap and CSV dataset of frequent transit stops using GTFS data, helping with resource allocation and service planning. Built with Scala and Spark for efficient data processing.
View Project
124,000 nodes and 68 million connections
A user-friendly knowledge graph that simplifies drug safety research by mapping connections and calculating similarity scores between drugs, genes, targets, and adverse effects, enhancing efficiency and decision-making in pharmacology.
View Project
A Python-based Django app for generating customer invoices with a built-in product and client database. Users can create profiles, add clients and products, select items from a dropdown when invoicing, generate PDFs for emailing, and mark invoices as paid.
View Project
A Django-based eBay-like auction site where users can post listings, place bids, comment on listings, and manage a personal watchlist.
View Project
Social Network is a Django-powered web application that mimics core features of Twitter, allowing users to create posts, follow others, like content, and manage social interactions—all within a dynamic and responsive interface.
View Project
This project is a front-end implementation of a single-page email client using JavaScript, HTML, and CSS. It interacts with a Django-based API to send, receive, read, archive, and reply to emails.
View Project
This is a Django web app that emulates core functionality of Wikipedia. Users can view, search, create, edit, and explore encyclopedia entries written in Markdown, which are then rendered into HTML for display.
View ProjectEmail: as.piyarali@gmail.com
GitHub: https://github.com/apiyarali