About Me
Hi, I'm Haoliang Cheng, currently a MS candidate in software engineering major at Carnegie Mellon University with a strong passion for AI application development, particularly in "to C" applications. I believe the potential for transformative AI solutions is immense, and I'm excited to contribute to this evolving field to create meaningful, innovative projects that improve people's lives.
I specialize in software development and machine learning model building, with experience applying AI in education and healthcare. I'm proficient in Python, Java, JavaScript, and also have experience in C, SQL and PyTorch, and I'm eager to leverage these skills in dynamic, impactful AI-driven solutions.
I'm always looking for new challenges and opportunities to grow both personally and professionally, with a focus on creating impactful AI-powered solutions that make a difference in people's lives.
Blog
Why Your LLM Hallucinates on PDFs' Images And How to Fix It
Published: January 16, 2026
Diving deep into why large language models struggle with PDF image processing and practical solutions to improve accuracy and reliability when handling visual content in documents.
Read MoreMy Understanding of Building AI/LLM Products
Published: July 23, 2025
Exploring the principles, challenges, and best practices for developing AI and large language model applications in the modern tech landscape.
Read MoreProjects
Paper Reading Agent
An AI-powered research paper reading agent that uses Google Gemini 3.0 to guide users through academic papers via a conversational workflow. Features PDF upload, smart summarization, interactive chat, on-demand figure extraction, and multilingual support, making complex academic literature more accessible and digestible.
💻 View Code on GitHub
CS Learning Path
A comprehensive platform dedicated to providing structured computer science and AI learning paths for self-learners, enabling everyone to master essential skills through carefully curated educational resources and AI-powered tutoring.
🌐 Visit CS Learning Path
My EEG
Using ensemble learning method to train and optimize machine learning model to predict human sleep stages and sleep quality with electroencephalogram (EEG) data.
💻 View Code on GitHubGreenhouse Agent
Successfully grew radishes and lettuce using an autonomous agent, running Python code on VM with Raspberry PI and translating data for Arduino in ROS to control the agent.
🎥 Watch Demo on YouTube