Taik Wang
Thousand Oaks
Summary
Fullstack Software Engineer with a focus on machine learning applications and deployment. Held a SECRET security clearance during 2021 - 2025.
Overview
4
4
years of professional experience
Work History
Software Engineer
Discovery Machine Inc.
Williamsport
03.2021 - 05.2025
- Engineered java based codebase for virtual instructor pilot training program, elevating simulation realism and performance stability.
- Simulated virtual test flights to evaluate system behavior, collected performance data, and submitted detailed, reproducible bug reports to streamline debugging and reliability.
- Collaborated with QA teams to track issues in Jira, facilitating timely resolution in subsequent builds.
- Conducted thorough code reviews and managed code integration using SVN, adhering to strict coding standards and stability across the development pipeline (CI/CD).
- Developed and maintained machine learning powered language services to capture communication used during our training program.
- Fine-tuned transformer based Automatic Speech Recognition (ASR) models, reducing domain specific WER from 23.4% → 6.2% (~73% improvement).
- Implemented a multi-voice Text-to-Speech (TTS) system to support varied simulation audio/languages.
- Built lightweight Python web applications to streamline user interaction with internal ML and training services.
- Utilized Docker Desktop to containerize the services to facilitate easy setup across different machines.
- Built production-ready, deployable software packages for on-premise software systems.
Education
Bachelor of Science - Computer Science
University of California, Irvine
06-2020
Skills
- Python
- Java
- C/C
- Docker
- Data Structures
- Machine learning
- MySQL
- Linux/Unix
- Version control
- SVN
- HTML
- Embedded Systems
Projects
- Rudimentary Audio Player, 03/01/20, Designed from scratch, a functional audio player capable of playing and recording songs. Utilized microprocessors and programming in C to record, process, and replay songs.
- Deepfake Learning Research, 04/01/20, 05/01/20, Created a fake image generator using PyTorch neural networks; generated images have uncanny resemblance to human drawn pictures.
Timeline
Software Engineer
Discovery Machine Inc.
03.2021 - 05.2025
Bachelor of Science - Computer Science
University of California, Irvine