Eng-Shen (James) Tu

About Me Updates Education Publications Research Experience Work Experience Past Projects Awards and Honors Other Activities
Email: jamestu6301@gmail.com

LinkedIn: Eng-Shen Tu

GitHub: JamesTu-jtjt

About Me

James

I am Eng-Shen (James) Tu, a Software Engineer at Appier and a research intern in the Software Engineering Group at Cornell University, advised by Assistant Professor Saikat Dutta. I earned my B.S. in Computer Science and Information Engineering from National Cheng Kung University (NCKU), Taiwan.

My work centers on AI for Software Engineering, including agentic and retrieval-augmented workflows, evaluation for LLM-based software agents, and automated software testing. I develop systems that improve reliability, software quality, and developer productivity.

You can reach me anytime or connect with me on LinkedIn.

CV

Research Interests:

  • AI for Software Engineering
  • Agentic LLM Workflows & Benchmarking
  • Retrieval-Augmented Reasoning
  • Automated Software Testing & Reliability
  • Human-Centered AI for Developers

What I do when I'm not coding:

Volleyball, Music, Chess, Casual Model UN

Explore my Model UN Experience

Updates

I am currently seeking research and PhD opportunities focusing on: AI4SE, Retrieval-Augmented and Agentic AI Workflows, and Human-Centered AI Systems for Software Development.

News

  • November 2025:
    • Invited as keynote speaker for the Statistics Department at NCKU Alumni Homecoming.
    • My Information Retrieval Research Group at the NYCU GDG has officially kicked off!
  • October 2025: Promoted at Appier from Intern to full-time Software Engineer (QA Automation) for AIRIS.
  • September 2025: Acceped invitation to join GDG @ NYCU as Project Leader once again.
  • August 2025: Won 2nd Place (Practical Impact Award) at Appier Hackathon for building QAgent.
  • July 2025: Returned to Appier as Quality Assurance Intern for BotBonnie.
  • June 2025:
    • Began work on OmniCode at Cornell University, a benchmark for software development agents.
    • CoverIQ has released 2 projects: Figma UI Test Planner and Local Unit Test Support!
    • CoverIQ won 3rd Place at the TCSE Software Engineering Competition!
  • May 2025:
    • Recruited engineering team for CoverIQ, welcoming Hank Chen(UIUC), Rex Yeh(NYCU), Claire Lin(Keio), and Wilson Liang(Columbia) onboard.
    • Selected for Cornell SE Group Summer Research Program under Prof. Dutta.
  • April 2025: Completed mandatory military service (53rd Engineering Group).
  • January 2025: JISE journal paper published on input relation prompting.
  • December 2024: Entered Taiwan mandatory military service.
  • August 2024: Appointed Project Leader for the LLM Research Group for GDG @ NYCU.
  • July 2024: Presented COMPSAC 2024 paper in Osaka, Japan.
  • May 2024: Led team to 2nd Place + Popularity Award at the Taipei Metro Hackathon.
  • November 2023: Selected for the government-funded TCA International Talent Program (Top 1%). Completed internship at SEAMEO RECSAM in Malaysia; received Global Supernova Award.
  • June 2023: Won Best Paper Award at TCSE 2023.
  • May 2023: Completed senior project on RLAIF for code generation (45% improvement across 145 tasks).
  • December 2022:
    • Won 2nd Place globally in IEEE Big Data Cup.
    • First-author IEEE Big Data paper published.
  • October 2022: Won MediaTek Better Future Award at NASA SpaceApps Challenge for multimodal text-to-image system.
  • 2020: Built Class Info Collector CLI for NCKU; deployed for internal use by Office of International Affairs.

Education

National Cheng Kung University (NCKU), Taiwan

B.S. in Computer Science and Information Engineering, 2019–2024

Publications

  • Anonymous, Tu E.-S. et al. “OmniCode: A Benchmark for Evaluating Software Development Agents.” Submitted to ICLR 2026.
  • Tu E.-S., Lee S.-J. “Input Relation Prompting for Metamorphic Testing on Query-Based Systems.” JISE, 41(1), 2025.
  • Tu E.-S., Lee S.-J. “A Clustering-Based Approach for Detecting Low-Contrast Texts on Web Pages.” IEEE COMPSAC 2024.
  • Hsu C.-C., Jian C.-Y., Tu E.-S. et al. “Real-Time Compressed Sensing for Hyperspectral Images.” IEEE TGRS, 2024.
  • Tu E.-S., Chen Y.-H., et al. “Rule-based Trip Destination Prediction.” IEEE BigData, 2022.
Google Scholar ORCID

Research Experience

Research Intern — Cornell University, Software Engineering Group

Advisor: Assistant Professor Saikat Dutta
June 2025 – Dec 2025

    Contributed to OmniCode (formerly CodeArena), recipient of the Meta AI LLM Evaluation Research Grant 2025. Omnicode is an an enhanced LLM Software Development Agent benchmark supporting Python, Java, and C++ across bug fixing, test generation, review response, and style fixing.
  • Implemented multilingual support, dataset construction workflows, and task evaluation pipelines.
  • Collaborated on benchmark infrastructure design, agent evaluation tooling, and integration pipelines.

Researcher — NCKU Computer and Network Center

Advisor: Associate Professor Shin-Jie Lee
Mar 2022 – Jul 2024

  • Conducted research on automated software testing, metamorphic testing, and SE quality assurance.
  • Proposed input relation prompting for oracle-less testing (Best Paper Award at TCSE 2023; JISE 2025 publication).
  • Designed a DBSCAN-inspired UI accessibility testing method achieving 80% precision (IEEE COMPSAC 2024).
  • Developed and maintained 20+ Scrapy-based web crawlers for industry collaboration projects.

Undergraduate Researcher — NCKU CSIE

Advisor: Professor Hung-Yu Kao
Sep 2022 – Aug 2023

  • Proposed an iterative RLAIF framework for code generation, improving weak model pass rates by 45% on 145 tasks.
  • Co-developed a space-themed text-to-image + style-content fusion system for NASA imagery (MediaTek Better Future Award, 2022 NASA SpaceApps).

Part-Time Research Assistant — NCKU Institute of Data Science

Advisors: Associate Prof. Chih-Chung Hsu; Professor Cheng-Te Li
May 2022 – Feb 2023

  • Worked on real-time compressed sensing for hyperspectral imaging; co-author on IEEE TGRS 2024.
  • Developed a rule-based system for trip destination prediction (2nd Place globally, IEEE Big Data Cup 2022; first-author publication).

Work Experience

Software Engineer — Appier

Jul 2025 – Present • Taipei, Taiwan

  • Designed a self-healing UI Testing Agent pipeline using Web UI agents + MCP for resilient E2E testing.
  • Built QAgent, an AI + RAG-powered test planning system adopted company-wide (2nd Place at 2025 Appier Hackathon).
  • Mentored 3 QA/SDET interns.

Taiwanese Army — Mandatory Military Service

Dec 2024 – Apr 2025

  • Served in the 53rd Engineering Group.

Quality Assurance Intern — Appier

Nov 2023 – Aug 2024 • Taipei, Taiwan

  • Supported test automation and regression testing across 19 Agile sprints.
  • Maintained automated pipelines in Jenkins and GitHub Actions.
  • Reduced manual QA workload by up to 50% through automation initiatives for AI-driven SaaS products.

Software Development Intern — SEAMEO RECSAM

Sep 2023 – Nov 2023 • Penang, Malaysia

  • Led full-stack development for a makerspace website.
  • Selected for the government-funded TCA International Talent Program and received the Global Supernova Award (Top 1%).

Past Projects

CoverIQ — AI-Powered Software Testing Solutions

2025 • Team Lead

  • Built Figma UI Test Planner (LLM + Figma parsing → BDD-style test cases).
  • Built Local Unit Test Support (LLM + RAG + AST for automated unit test generation).
  • Interviewed senior QA/SDET engineers across Appier, GoFreight, Tomofun; secured prototype trial interest.

Herewegoal — Lightweight Productivity Tool

2025 • Product Manager

  • Designed a minimalist project tracking interface supporting collaborative workflows.
  • Contributed to product planning, operations, and quality assurance.

MRT Pet Project — 2nd Place @ Taipei Metro Hackathon

2024 • Team Lead

  • Proposed an AI-personalized MRT companion app to improve rider engagement.
  • Led team of 5; recognized among 90+ teams for innovation and usability.

RLAIF for Code Generation — Senior Project

2023

  • Achieved 45% pass-rate improvement on 145 Python coding tasks through RLAIF fine-tuning.

JTHankAI — NASA SpaceApps Hackathon

2022 • Team Lead

  • Developed a multimodal text-to-image + style-content fusion system for NASA’s imagery dataset.
  • Won the MediaTek Better Future Award.

Class Info Collector — CLI Tool

2020

  • Scraped and organized university course data; deployed internally to NCKU Office of International Affairs.

Awards and Honors

  • Practical Impact Award — 2nd Place, Appier Hackathon (2025)
  • 3rd Place, TCSE Software Engineering Competition (2025)
  • Second Place & Popularity Award, Taipei Metro Hackathon (2024)
  • Global Supernova Award, TCA International Talent Program (2023)
  • Best Paper Award, TCSE (2023)
  • Second Place, IEEE Big Data Cup (2022)
  • MediaTek Better Future Award, NASA SpaceApps Challenge (2022)

Extracurricular Leadership

Google Developer Group OnCampus — NYCU

Project Leader, LLM Research Group (2024–2025)

MUN Society Taiwan

Director (2021–2025)

Harvard Undergraduate Taiwan Leadership Conference

Head Teaching Assistant (2022, 2024)

NCKU Statistics Department Women’s Volleyball Team

Head Coach (2022–2024)

Harvard WorldMUN Conference

Assistant Chair (2024)

NCKU Blockchain Club

Director of Research (2021–2022)

Hult Prize at NCKU

Founder & Campus Director (2020–2021)