Data Scientist, Engineer, AI, Educator
This project is a hands-on implementation of a real-time algorithmic trading system that processes market data and news sentiment to make automated trading decisions. It integrates Apache Flink for stream processing, Redpanda for data streaming, and the Alpaca API for market data and trade execution, with sentiment analysis driving the strategy. Technologies Used How It …
This project is a real-time anomaly detection system for stock trading data, built with Redpanda and Quix. It processes streaming trade data and flags unusual patterns using both rule-based and machine learning techniques. Purpose The goal is to detect anomalies—like sudden price jumps or high-volume trades—as they happen, using a lightweight, practical setup. Technologies Used …
In today’s data-driven world, real-time monitoring of database changes is critical for applications like fraud detection and live analytics dashboards. This project demonstrates a robust real-time data pipeline using PostgreSQL, Debezium, and Kafka, orchestrated with Docker, to capture and stream database changes efficiently. The project includes: Features Technology Used Learn More Link: Read the full …
This repo shows how to use AWS EMR and Apache Spark to process big data in the cloud. It includes scripts and notes from my Medium guide on setting up and running Spark jobs. Key Features Tech Stack Link: Read the full article on Medium Code: https://github.com/shj37/AWS-EMR-Scalable-Spark-Processing
This project demonstrates the creation of an employee churn prediction pipeline using Google BigQuery, PyCaret, and Looker Studio. The goal is to predict which employees might leave based on historical data and offer insights to boost retention. Technologies Used Key Features Link: Read the full article on Medium Code: https://github.com/shj37/Employee-Churn-Prediction-with-Looker-Studio-BigQuery-and-PyCaret
Summary Skills: Fine Tuning · Large Language Models (LLM) · Natural Language Processing (NLP) · TPU · GPU · Artificial Intelligence (AI) · LoRA · QLoRA · FinTech · financial risk compliance · Chatbot Development · Keras · PyTorch · Project Management · Gemma-2b · Gemma-7b · Data Cleaning · Synthetic Data Generation Highlights Project …
Role: Student AI Researcher at Skinopathy Skills: Artificial Intelligence (AI) · speech-to-text · Fine Tuning · QLoRA · Natural Language Processing (NLP) · Text-to-Speech Synthesis · Data Processing · Data Cleaning · OpenAI Whisper · Audio Recognition · Project Management
I conducted some preliminary tests to assess the efficacy of employing GPT-4 for analyzing reference documents and generating answers for tasks assigned in the first semester of my graduate certificate program. It’s worth noting that this project was carried out post-completion of those courses, and therefore, was not utilized for the assignments associated with those …
Recently, I discovered that even with the above comprehensive schedule encompassing all class choices, creating diverse timetables to meet various preferences can be time-consuming. For instance, some of my friends prefer not to have classes on Saturdays, others aim to condense classes into fewer days, and I personally want to avoid morning sessions. Additionally, there …
In the realm of online shopping and entertainment, we often encounter personalized recommendations like “You may also like this item” or “Recommended movies based on your preferences.” These suggestions stem from association rule mining algorithms, specifically market basket analysis. This method uncovers item relationships from historical data, shedding light on customer behavior and preferences. Recently, …