# Wenjun S.

> Software Engineer at Everlaw

Location: San Francisco, California, United States
Profile: https://flows.cv/wenjuns

I am passionate about using technology to better our daily lives. I am specifically interested in AI, education, open source projects, and working with non-profits. Currently, I am working as a full stack engineer at Everlaw.

## Work Experience
### Software Engineer @ Everlaw
Jan 2022 – Present | Oakland, California, United States
- Full stack engineer working across the tech stack to ship high value features.
- Tech stack includes MySQL, Spring, Java, React, TypeScript

### Machine Learning Researcher @ Ubiquitous Computing Lab
Jan 2021 – Jan 2022 | Seattle, Washington, United States
Worked with Chunjong Park at UW Ubicomp Lab on machine learning fairness
- Conducted experiments in PyTorch to detect bias in deep learning systems for a variety of data sources.
- Used explainable AI methods such as GradCAM to inspect what a biased deep model is learning.
- Proposed new methods using OOD (out-of-distribution) scores to help debiasing deep learning models.

### Undergraduate Researcher @ Ubiquitous Computing Lab
Jan 2021 – Jan 2022 | Seattle, Washington, United States
Worked with Jason Hoffman at UW MISL and Ubicomp Lab on DNA technology and virus detection
- Built a software pipeline using DNA analysis packages such as NUPACK and KINDA so that given an virus DNA/RNA sequence, the system can output multiple DNA circuit designs that can potentially be used to detect virus in a biological lab setting.

### Undergraduate Researcher @ Ubiquitous Computing Lab
Jan 2020 – Jan 2021 | Seattle, Washington, United States
Worked with Chunjong Park at UW Ubicomp Lab on COVID 19 behavioral change causal analysis
- Preprocessed and sanitized gigabytes of behavioral data collected from people's mobile phones during COVID.
- Visualized people's behavioral change time series using Matplotlib. 
- Utilized causal inference approaches such as diff-in-diff analysis, propensity score matching to find out whether government's shelter in place policy causes people to stay at home more.
- Used Double Machine Learning, Causal Tree, and SHAP values to find out heterogeneity in causal effects, that is, why some counties have more behavioral change than other counties.

### CSE 446 (Machine Learning) Teaching Assistant @ Paul G. Allen School of Computer Science & Engineering
Jan 2022 – Jan 2022 | Seattle, Washington, United States
- Held weekly office hours to help students on homework questions
- Regularly monitored Ed discussion board to answer students' questions about the course
- Graded homework and provided detailed feedback to students
- planned the section materials and homework questions

### Software Engineer @ F5
Jan 2021 – Jan 2021 | Seattle, Washington, United States
I helped my team collect relevant HTTP performance data using bash scripts and command line tools. 
Then I built a data dashboard using Python Dash library and presented insights from the data to my team.
Also, I wrote code to test the new API schema change that our team was working on, and used docker containers to validate that our schema change indeed worked.

### Software Developer @ UW Impact++
Jan 2020 – Jan 2021 | Greater Seattle Area
UW Impact++ is a student lead organization at UW that aims to empower non-profit and open source organizations with technology.
During my time at Impact++, I specifically worked on open source project called ODK that supports free mobile data collection.
Specifically I used Javascript to make a Google Data Studio connector to the ODK database so users can visualize their data online with ease.
Our GDS connector has been reviewed and published by Google, and it is now an official data source that ODK users can use to explore their data!

### Undergraduate Researcher @ University of Washington
Jan 2020 – Jan 2021 | Seattle, Washington, United States
Worked with Emisa Nategh on applying machine learning to a medical problem.
	• Did feature selection based on correlation, mutual information, chi-square, and random forest scores.
	• Utilized K-Mode clustering algorithm on categorical feature vectors to group similar patients together
	• Used T-SNE algorithm to visualize high dimensional data in 2-D and 3-D.
	• Used AutoML from H2O AI platform to quickly train machine learning models with the best hyperparameters to predict a patient condition.


## Education
### Bachelor of Science - BS in Computer Science
University of Washington

### Java Web Developer Nanodegree
Udacity

### High School Diploma
Xavier College Preparatory High School


## Contact & Social
- LinkedIn: https://linkedin.com/in/wenjunsun

---
Source: https://flows.cv/wenjuns
JSON Resume: https://flows.cv/wenjuns/resume.json
Last updated: 2026-04-11