#!/usr/bin/python
# -*- coding: utf-8 -*-
class SoftwareEngineer:
def __init__(self):
self.name = "Jeffrey Liu"
self.role = "Software Engineer"
self.language_spoken = ["en_US", "zh_CN"]
self.interests = ["Stand-Up Comedy", "NBA", "Ice Hockey", "Poker", "Anime (Attach on Titan)", "Go", "Chess"]
def say_hi(self):
print("Thanks for visiting my personal Github Repo, hope you find some of my featured work interesting.")
me = SoftwareEngineer()
me.say_hi()
- Personal Website: Website Link
- 🚀 Incoming SDE Intern @LinkedIn (Summer 2024)
- 🔭 Ex SDE Intern @Amazon (Summer 2023)
- 🌱 I’m currently learning about Nature Language Processing and LLM Finetuning.
- 💬 Feel Free to connect with me. I love communicating with everyone.
- 😄 Pronouns: He / Him / His
A Job Dashboard and Newsletter website that helps student to find their best-fit internship. Exciting Update! The Project now has 200+ active users.
What makes exact-match.io a special project? 🛡️
- 📨We have a Weekly Newsletter feature! Get tailored notifications about job postings that are in sync with your profile - your Grade Level, Sponsorship needs, and more.
- 🔍 Our intuitive homepage lets you Select and Filter job postings based on your criteria. The ‘Apply’ button will whisk you away directly to the job posting page. 💨
- A large language model (GPT) supported Positive Reframing Website, allowing users to restructure sentences into more positive phrasing while retaining the essence of the content.
- Integrated the HuggingFace Gradio API and ChatGPT API into the system, allowing users to receive multiple rephrased sentence options for selection.
Paper Title: Did your child get scared by an inappropriate advertisement on YouTube? Website
- First Author
- Led a team of four student researchers to construct a large-scale and labeled dataset of YouTube advertisements displayed in young-kid-oriented videos.
- Accepted by 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining Undergraduate Consortium (KDD-UC).
- Tech: Selenium, Python, Web Scraping, HTML (May '22)
Paper Title: Samba: Identifying Inappropriate Videos for Young Children Website
- Accepted by 31st The Conference on Information and Knowledge Management (CIKM)
- Collaborated with researchers and professors to develop a fusion model that improves the accuracy of predicting whether a YouTube video is appropriate for children by at least 7% over other state-of-the-art classifiers.
- Tech: Tensorflow, BERT (May '22)
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent