I am a dedicated and passionate software developer with a strong background in web development, specializing in creating responsive, user-friendly, and visually appealing experiences. Skilled in HTML, CSS, JavaScript, and frameworks like Bootstrap, I continuously explore new technologies and industry trends to enhance my skills. My interest in artificial intelligence and machine learning drives me to apply AI algorithms in web development, developing intelligent, data-driven solutions. Particularly interested in local LLMs which drive me to experiment with different local models such as Qwen 3, DeepSeek, gpt-oss, and others. Additionally, I have a solid foundation in system-level programming, including operating systems, multithreading, and performance optimization. I thrive on collaboration, problem-solving, and adapting to new challenges, ensuring I deliver high-quality, modern solutions. In my free time, I enjoy working on personal projects and contributing to open-source communities.
Yuri Fung
I'm a
About
Computer Scientist, Software Engineer,
Web Developer, Data Analyst
Currently, I am attending the University of Colorado Boulder.
I am majoring in Computer Science. I just finished my Bachelor's and I am pursuing a Master's in Computer Science with a focus in AI/ML.
Right now, I am currently working on a personal research paper to run Large Language Models on regular phones at 30 tokens/sec or higher. Furthermore, some side projects that will be listed below soon.
Online Profiles
- Website:superintiald.github.io
- LinkedIn: yuri-m-fung
- GitHub: SUPERINTIALD
- X:@yur1chantan
- Facebook:yur1chantan
- Instagram:@yurichantan
- LeetCode: SuperInitialD
Personal Information
- Major: Computer Science
- Degree: Bachelor's & Master's
- Email: fung.yuri.intern@gmail.com
- State: Colorado, USA
Portfolio
.png)
PerfectHomeFinder
Real-time home searching app using AI and data-driven APIs
Welcome to PerfectHomeFinder!
Example using monte carlo and sarima
Example using data/numpy
Example using Pandas
Discover your dream home with our advanced search features.
Explore properties with detailed information and high-quality images.
Find homes that match your preferences and budget.
AutoDS (Agentic Data Science)
Formally known as Data2Paper, this project began as an automated data-to-paper solution for research professionals. Initially, we used Data2Paper as a framework, but later shifted from this ideology to develop a more reasoning-based model, where each agent communicates with APIs and LLMs to minimize hallucinations. The release of Deepseek, a reasoning LLM model, was pivotal in advancing our project. Leveraging LangChain and LangGraph, we built an agentic model that connects a network of LLMs, ultimately creating an AutoDS framework designed to assist data scientists.
Watch the project demo playlist.
Data2Paper automates the process of converting data into research papers.
Agentic Data Science Project Expo @CU Boulder
Our mascot came to visit us!
Final Poster for Agentic AI for AutoDS/AIDS
Streamline your research workflow with our innovative tools.
Generate high-quality research papers from raw data effortlessly.
Example 1: Generate high-quality research papers from raw data effortlessly.
Example 2: Generate high-quality research papers from raw data effortlessly.
Metanoia - 2024 Hackathon Project
Chrome extension for journaling and personal reflection, awarded 1st place in UI/UX category.
Metanoia helps you reflect on your day and improve mental well-being.
Metanoia helps you reflect on your day and improve mental well-being.
HackCU 2024 Participants.
SportSphere
Sports betting tracker with user authentication and live sports data.
Access all features from the intuitive and user-friendly homepage.
Track your betting history and analyze your performance.
Get real-time sports data and make informed betting decisions.
Join the community and discuss betting strategies with other users.
UML of Sportsphere app
Explore the wireframe design of the SportSphere app.
Figma Design
Design and prototype with Figma.
Website icon
Collaborate with your team in real-time.
Streamline your design process with Figma.
Explore the human-computer interaction design with Figma.
Bowling
Bowling is one of my favorite hobbies. I enjoy the challenge and precision required to consistently improve my game. It's a great way to relax and have fun with friends.
Enjoy a fun game of bowling with friends and family.
Stay fit and healthy with regular exercise.
Relax and unwind with your favorite hobbies.
Data Mining Project - Housing Price Prediction
Implemented SARIMA, ARIMA, and additional machine learning models to predict housing prices in Boulder, CO, leveraging detailed data pre-processing, feature engineering, and model evaluation techniques.
Data preprocessing and feature selection. Collected housing price data and key attributes such as property size, neighborhood details, and economic factors. Ensured data cleaning and normalization.
Data preprocessing and feature selection. Collected housing price data and key attributes such as property size, neighborhood details, and economic factors. Ensured data cleaning and normalization.
ACF and PACF analysis. Autocorrelation Function (ACF) and Partial Autocorrelation Function (PACF) were used to determine lag terms and seasonality in the time series. This helped configure ARIMA and SARIMA models effectively.
Residual analysis for model validation. Checked residuals to ensure they were normally distributed and exhibited no significant autocorrelation, indicating a well-fitted model. Performed Ljung-Box tests to validate assumptions.
ARIMA model performance evaluation. Tested various configurations of ARIMA to identify optimal lag, differencing, and seasonality parameters. Explored ACF and PACF plots for parameter tuning.
SARIMA implementation. Incorporated seasonality to capture cyclical trends in Boulder’s housing market. Evaluated performance based on RMSE and MAE metrics, with seasonality accounting for quarterly fluctuations.
HackCU2025 - Hermes
Chrome extension for journaling and personal reflection, awarded 1st place in UI/UX category.
Hermes provides a streamlined platform for managing organizational tasks efficiently.
Access detailed information about your organization and its members.
Learn more about Hermes and its features to enhance productivity.
Organize and track events seamlessly with Hermes' event management tools.
Manage fraternity or similar organizations with tailored features for collaboration.
Streamline organizational workflows and improve team coordination.
Explore the system architecture that powers Hermes' robust functionality.
OOP - Research and Game Project
Created a Polymorphia Game in Java, and researched Python vs Java in Game Devlopment
Created an API server for Polymorphia
Polymorphia in game
UML Polymorphia
Upgraded UML Version Polymorphia
OOP - Research and Game Project
Created a Polymorphia Game in Java, and researched Python vs Java in Game Devlopment
Interns of 2025
Name plate :)
Worked in Cupertino California!
Click on a project image above to view more details.
Featured Profiles & Repositories
Research Projects
Articles
Teaching a transformer to recognize car brand, model and engines from sound
A practical write-up describing training a transformer to classify car characteristics from engine audio. Walks through dataset preparation, model choices, and results.
Beyond Chatbots: Efficient Embodied Reasoning and the Path Towards Proactive Agents
A deep dive into agentic systems that combine embodied reasoning, efficient model orchestration, and proactive behavior to move past traditional chatbot limitations.
What Actually Improved Push-T
An exploration of the advancements in the Push-T model, focusing on the key changes that led to its improved performance.
Resume
A highly motivated and detail-oriented Computer Science student at CU Boulder with experience in software development, cybersecurity, and machine learning.
Education & Honors
University of Colorado Boulder, College of Engineering & Applied Science
Master's Degree in Computer Science (BS)Expected Graduation: May 2026
- Accepted into Bachelor’s Accelerated Masters (BAM) Program
- Fall 2024 Coursework: Data Mining, Object-Oriented Programming
Bachelor's Degree in Computer Science (BS)Graduated: May 2025
- GPA: 3.9
- Coursework: Operating Systems, Algorithms, Database Systems, Data Science, Linear Algebra, Data Structures, Computer Systems, Software Development, Introduction to AI, Machine Learning, Principles to Programming Languages, Introduction to Robotics, HCI, Discrete Math, Calculus (1,2,3), Cyber Security, Capstone Software Development Project (Sponsored by Trimble)
- Awards:
- National Society of Leadership and Success Honors
- Dean's List: Fall 2022 - Spring 2024
Professional Experience
MOVE IRA Reviewer Independent Contractor
Handshake, 08/2025 - Present
- Promoted From Fellow to IRA Reviewer in a role based on strong performance and accuracy
- Evaluated Undergrad, Graduate, and PhD level tasks designed to challenge AI systems
- Assessed outputs for accuracy, reasoning quality, and compliance with task requirements.
- Provided structured feedback to improve system reliability and correctness
- Identified document reasoning failures, extraction failures, and ambiguities in model behaviors
MOVE Fellow Independent Contractor
Handshake, 08/2025 - 09/2025
- Created undergraduate- and graduate-level tasks to test model robustness.
- Reviewed model responses and contributed to baseline evaluation benchmarks.
- Gained experience in AI evaluation and quality assurance, leading to promotion.
Software Engineer: Wireless Technology and Ecosystems
Apple, 05/2025 - 08/2025
- Developed software solutions using Objective-C, C++, and Swift across 4 iOS frameworks.
- Built proof-of-concept implementations across telephony and messages and presented solutions to engineering teams, managers, and directors.
- Contributed to connectivity features with potential to impact 1B+ users through improved reliability.
- Pitched an accessibility solution in Apple's intern iContest.
- Collaborated on LLM-driven tools to streamline internal workflows, which contributed to data analysis and decision making.
Software Developer
Technical Consulting and Research (TCR), 02/2025 - Present
- Collaborating on full-stack web development for government websites—including hospitals—by designing, validating, and securing platforms using advanced authentication (e.g., Auth0), tenant architectures, API integrations, and adhering to cyber security and coding best practices.
- Actively engaged in a dynamic team project, mapping out flow charts, exploring both front-end and back-end systems, and meeting weekly to strategize and update progress.
Cashier Assistant
Costco, 07/2022 – Present
- Worked a variety of roles including cashier assistant, self-checkout, carts, stocking, bakery, deli, and more.
- Delivered an unparalleled customer experience, positively affecting customer satisfaction and retention daily.
Chef/Food Prep
Piripi, 06/2021 – 07/2022
- Prepared Caribbean/Mediterranean dishes and coordinated with the kitchen team to ensure smooth operations.
Leadership and Activities
Team Lead, Software Development Project
01/2024 – 05/2024
- Led a team in an Agile project for a software development initiative, focusing on task delegation and milestone achievements.
Hackathon Leadership
Managed and coordinated teams in hackathons, including HackCU and T9, achieving top placements through effective time management and technical problem-solving.
Projects
PerfectHouseFinder, Self-employed Startup
09/2024 – Present
- Launched a self-employed startup focused on helping users find their ideal home based on personalized queries. Integrated diverse datasets and external APIs to provide real-time information on crime rates, school quality, geography, home prices, neighborhood characteristics, and local amenities.
- Implemented Natural Language Processing (NLP) using models such as GPT-2, GPT-4, GPT-4o, Llama-3, Llama-3-8b-Intruct, DialoGPT, Roberta-base-Squad-2, and GPT4All to enhance search accuracy and user interaction.
Software Engineering Project, Trimble (Capstone) – Technical Team Lead
09/2024 – Present
- Led a team in developing a data-to-paper solution for Trimble, focused on minimizing hallucinations in generated content and enhancing novelty and creativity in research output.
- Modified existing algorithms and incorporated an additional step from data-to-paper-to-presentation, enabling Trimble to efficiently identify and present research topics. Leveraged Meta-GPT and LLMs such as GPT-4o, and utilizing Semantic Scholar and A/synchronous OpenAI API keys to improve the quality of generated content.
OOAD Polymorphia Project, Team-lead
09/2024 – Present
- Developed Polymorphia, a maze-based adventure game in Java using IntelliJ, incorporating object-oriented design for dynamic gameplay with adventurers, creatures, and combat mechanics.
- Researched the difference between Python and Java for OOP-based game development/design.
Cybersecurity Labs
10/2024 – Present
- Identified and exploited web vulnerabilities, including XSS (stole cookies) and SQL Injection (stole admin credentials), using tools like Inspect Element (developer tools), Burp Suite, Bandit, and fuzzing for dynamic and static analysis. Conducted threat modeling and tested web clients and servers, then patched all vulnerabilities to enhance security.
- Incorporated Nmap, Hashcat, Wireguard, and Wireshark to exploit and find vulnerabilities in web servers.
D4D, Designing for the Department of Defense (DoD)
01/2024 – 05/2024
- Collaborated with the DoD to incorporate machine learning into projects through the Air Force and Navy.
- Air Force Project: Incorporated automation for refueling operations through the F-15 to enhance efficiency without human intervention.
- Navy Project: Developed an extended-range system for UUVs, enhancing deployment and retrieval processes through machine learning and deep learning.
SportSphere, Team-Lead
01/2024 – 05/2024
- Led the development of a sports betting tracker website, implementing user authentication, backend API integration for live sports data, and database management using SQL.
- Collaborated on front-end pages (home, user, betting, sports, community, about) to display real-time events and user bets, while finalizing key features like community-based tracking and a leaderboard system.
HackCU, Metanoia
03/2024
- Placed 1st in UI/UX category for a 24-hour hackathon. Worked on the front and back-end of this Chrome extension, Metanoia, a digital time capsule allowing users to journal responses to thought-provoking prompts that cycle annually, enabling users to reflect on personal growth over time.
T9 Hackathon, Clothin’ in a Can
02/2024
- 24-Hour Hackathon project: Designed a sustainable, compactable, and biodegradable fashion solution called "Clothin' in a Can."
Multi-threaded DNS Resolver
10/2023 – 11/2023
- Developed a multi-threaded DNS resolver in C, utilizing semaphores and mutexes to efficiently map domain names to IP addresses using requester and resolver threads, processing multiple input files simultaneously with a thread-safe shared array.
Attack Lab
03/2023
- Used assembly code to perform five buffer overflow attacks, including three code injection attacks and two return-oriented programming attacks on executable files.
Performance Lab
03/2023
- Optimized a program to be approximately 1000x faster through multi-threading and signal masking, reducing the CPE from 3900-4000 to 4.
Mini Git
12/2022
- Recreated Git functionality in C++ using linked lists, stacks, queues, and hash maps. The project supports operations like git commit, git branch, git init, git add, git push, and git clone.
Skills
Technical Skills
- Programming Languages: C++, C, Java, Python (ISLP, Pandas, Numpy), Flask, Pytorch, TensorFlow, JavaScript, NodeJS, Handlebars, Assembly, LaTeX, Apple Script, R, SQL (NoSQL), Scala, HTML, CSS, XML, Shell Script, PHP, MATLAB, Lua
- Applications/Tools/Frameworks: Jupyter Notebook/Lab, Figma, Docker, Airflow, Postgres, MongoDB, Postman, JSON, AWS Bedrock, Titan, VS Code, IntelliJ, VMware, VirtualBox (Oracle), RStudio, PyCharm, Adobe Photoshop, Bash, Google OAuth, Microsoft Azure, Bootstrap, Git, Anaconda, CUDA, Burp Suite, Gradle, MySQL Workbench, Webot, iPhone, Android, SikuliX, Project Management (Agile, Waterfall), Version Control (GitHub), Wezterm, Wireshark, Wireguard, Nmap, Hashcat
- Other Skills: Robotics (VEXCODE), Web Development (Frontend, Backend, Full Stack), UI/UX Design, Data Modeling, Automation, OS Expertise (Windows, macOS, Linux), Google Suite, Microsoft Office, Google Workspace