About

Hi! I'm Wan. You can reach me at [email protected].

Random fun facts:

  • I believe that I'm the first (and only) Bruneian undergraduate PhD student at CMU.
  • I believe that I'm also the first CMU CS undergrad to graduate with two SCS concentrations, though that's mostly from being the first concentrations batch and stumbling my way into it.
  • I've helped to raise chickens. I also have a MOOC cert in chicken behavior.
  • I grew up in "starfruit village" -- that's where belimbings (fake plural) comes from. The singular version is taken by a domain squatter, so I have belimbings.com.

Interests

I like systems programming, distributed systems, and databases. I'm generally interested in the tradeoffs that people make between speed, memory, formalism, and communication. I admire efficient systems and reproducible research.

List of things that I want to learn more about some day, in no particular order: combinatorics, information theory, streaming algorithms, randomized algorithms, approximation algorithms. I used to be into cybersecurity, which is probably how I got into CMU as an undergrad (CTFs). I will be somewhat disappointed if "scale is all you need" nowadays.

Hobbies

I used to play DOTA2, TF2, and assorted MMORPGs. I also went through a brief League, CSGO, and Valorant phase. I don't play much nowadays.

I enjoy creating teaching material, typically aimed at an introductory level. I particularly believe that everyone should have access to good learning resources (whether they make good use of it or not is up to them). If you're doing something in this space, I'm interested in chatting! Please include "capybara" in your message to get through spam.

I spend an inordinate amount of my time sleeping, reading about food, or browsing cute subreddits. Some good YouTube channels:

Things I Use

This is not an explicit endorsement, but I use the following. If you have recommendations, please feel free to email me. I am aware that this is poor opsec.

Bookshelf

2022 comment: I don't remember when I last updated this section; I'm not in the mood to update it yet.

(Research papers not included, fiction not included.)

I personally prefer hardcopies of books and buy books that I consider to have high value. This list only contains books which I have bought or read significant parts of. Always happy to take book recommendations or reviews!

Currently reading:

  • TODO update this.

To be continued:

  • All of Statistics.
    I like the presentation of the book and hear good things about it.
    I'm dubious of appeals to nature and stuff like that in a number of introductory machine learning texts, so I am looking forward to reading this book's treatment of the topic.

  • The Shellcoder's Handbook.
    On my bookshelf overseas. Got me started with basic binary exploitation. Progress was slow in high school. Should be much faster now that I've written an operating system kernel.

  • Practical Malware Analysis.
    On my bookshelf overseas. Got me started with basic reversing and detection avoidance. Progress was slow in high school. Should be much faster now that I've taken 15-213.

To be read some day:

  • A Book of Abstract Algebra.
    On my bookshelf. I don't remember anything from taking Algebraic Structures and have no intuition whatsoever. Definitely was not ready for Dummit and Foote back then.

  • The Nature of Computation.
    On my bookshelf. Nicely written book from a new point of view (Moore is a physicist). Covers a mix of topics from algorithms and undergraduate complexity theory.

  • Mathematics: Its Contents Methods and Meaning.
    On my bookshelf. I have extremely shaky foundations in math, in particular my geometry is nonexistent. This seems like a decent way of shoring it up.

To be re-read:

  • Performance Modeling and Design of Computer Systems.
    The material didn't sink in the first time (15-359), and it didn't really sink in the second time either (15-857), so maybe the third time is the charm. To be clear, that's mostly a me problem. My goal is to be able to quickly estimate "X workload? Deploy Y servers in Z configuration".

Would love to read, doubt it will practically happen in the next decade:

  • Practical Foundations for Programming Languages.
    Didn't really read this enough while taking the class.
    I think formally verified things will be the future, just not "in my working lifetime" future.

  • A Textbook of Modern Toxicology.
    I was really into this book at some point. Good for going on Wikipedia binges.

Have read, can recommend:

  • A Teacher's Race Course.
    This book is a little different from the others on the list.
    It was written by a teacher of mine who has since passed away.
    She strongly influenced my teaching beliefs and guided me towards a more positive outlook of the world. A key part of her teaching philosophy is to treat and genuinely care for students as people first and foremost. She had a reputation for being able to turn even the most misbehaved and poor-scoring set of classroom monkeys into competent students. Anyone who wants to seriously improve their teaching would probably benefit from reading this book.

  • Poor Economics.
    This book gets it. I typically roll my eyes at most college student discussions of economics or poverty, but this book provides an understanding and empathetic view to questions like: "why don't the poor save more? why don't they invest in their future? why don't they use free mosquito nets? what interventions actually work?", and so on. Good and varied selection of case studies, engaging discourse. If I had read this at an earlier stage in my life, I might have gone into economics.

  • The Web Application Hacker's Handbook.
    On my bookshelf at home. Gentle introduction to exploiting web applications. I was able to follow along and directly apply its lessons to real-world web applications in high school.

Projects

I would like to work on these at some point. Inspiration is generally an issue.

    1. Falls are the leading cause of injuries (fatal and otherwise) for older people. I've envisioned a cheap wearable everyday fall-arresting device, targeted towards the elderly and infirm. I haven't been able to come up with a reasonable concept, though, and I'm bad at hardware problems. Dumb ideas have included weight shifting clothing, quick inflating swarm quadcopters, and airbag-esque pockets. You could also repurpose something like ReWalk, presumably, but that's way too expensive.
    1. (A little depressing) Recently, I've been thinking more about pagpag. Fundamentally, I believe that most people that I've met would be kind enough to give up their Starbucks or bubble tea or whatever it is if they could donate directly to people like those in this video. I don't really know what a solution would look like in this space -- can you use technology to reach and help people who fundamentally have zero access to technology? For less extreme cases, GiveDirectly is an interesting manifestation of this idea (households need at least SIM cards afaik), and ActionAgainstHunger is a generally safe choice as far as food charities go (93 cents on the dollar). Both are pretty well rated charities.