A bit more about me

As I stated in my home page, I build for impact. I come from a background dominated by machine learning for bioinformatics and healthcare applications.

I've been building fullstack applications for startups since December 2023. Prior to that I have been coding for five years since I was a freshman in high school in May 2018. I originally came from a machine learning (ML) background. Some of my earlier projects include a CNN-based oracle for diagnosing and classifying cardiac arrhythmias, a chest X-ray pneumonia classifier.

On research, I have collaborated on projects such as:

  • Detecting mutations from H&E biopsy slides of malignant tumors, with Prof. Aristotelis Tsirigos (NYU Medicine, Chair of Bioinformatics)
  • Detecting kidney transplant complications using LSTMs, with Dr. Christopher Nguan (UBC Medicine)
  • Computer vision to model pig brain segmentation, with Prof. Matthew Caesar (UIUC Computer Science)
  • This project was presented at the Swine in Biomedical Research (SBR) conference in Madison, WI in June 2022.

I am now open today on working on applied problems in a wide range of industries. I build products with these three principles that I achieve in this order:

  • Build it right. I make sure the application works reliably without bugs, and I set high standards for the code that I write. I try to minimize tech debt and write modular, documented. I very much enjoy taking a highly formal, top-down approach to writing my applications.
  • Build it well. I listen to users, get feedback on features and UI/UX, and iteratively improve my applications. At this point, the existing features are already reliable and performant.
  • Build it fast. By then, I will have much practice building high-quality applications. With modular, clear, well-documented code, working on focused problems, I become much faster at shipping features or even end-to-end applications. The amount of time I take to build and ship features then range from a few hours to a few days.

I find it much easier to achieve all three principles in this order than the other way around. I have unfortunately seen too many teams suffer from immense technical debt. I have seen promising startup ideas brought to screeching-halt feature velocity because they were servicing technical debt. While I am open to joining pre-established teams, I prefer to set technical standards as an early/founding engineer in a high-trust work environment.