whoami

A researcher based in Japan, from Binh Dinh

Education

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15


$ cat /proc/education

Masters:
  Institute: "Japan Advanced Institute of Science and Technology"
  Major: "Knowledge Science"
  ResearchTopics:
    - Knowledge Graphs
    - Explainable Machine Learning

Bachelors:
  Institute: "University of Science, Ho Chi Minh City"
  Major: "Computer Science"
  Range: [2019, 2023]
  ResearchTopics:
    - Machine Translation

For professional experience and publications, refer to my LinkedIn.

A Wanderer

While “researcher” might be the conventional label for my professional work, I’ve always seen myself more as a wanderer.

From tinkering with old computers, bricking and reviving Android devices during secondary school years, and running unlikely operating systems on unsupported hardware, I’ve carried a habit of exploring technology in slightly unconventional ways. None of my phone ran the original OS they originally came with (Apple devices are exceptions). That spirit of exploration continues to guide me, whether in research, hacking around with systems, or simply following curiosity wherever it leads.

Current toys

Main workstation: MacBook Air M1 (16/256)
Secondary workstation: Surface Pro 8 (8/256; allocated by the institute)
Toy NAS: Mac Mini Mid-2011 (4/512; Ubuntu)
Handheld console: Anbernic RG35XXSP
Camera: Canon PowerShot G7X
Boring things: iPad Pro M1, iPhone 16 Pro

Retired toys

Click to expand: PC

ThinkPad T14s (R5 4650U; 16/512; dualboot Arch Linux–Windows)
ThinkPad T430s (i5-3230M; 8/256; tri-boot macOS/Hackintosh–Arch Linux–Windows)
Potato i3-2100 PC (4/320)
Potato Pentium 4 PC (2/80)
Potato Pentium III PC

Click to expand: Gadgets

Symbian phones: Nokia 2700c (2011), Nokia X2-01 (‘13), Nokia Asha 303 (‘14)
Smartphones: Galaxy GT-S5300s (‘15), Zenfone 2 (‘15), Redmi 5 Plus (‘17), Redmi Note 9s (‘20), iPhone 13 (‘23)
Tablets: FonePad 7 (‘13), MatePad 10.4 (‘22)
Camera: PowerShot SX-400HS, PowerShot 〇〇 (don’t remember)

What’s ‘hiraki’?

It’s the kunyomi reading of my name 發 in kanji. A linguistically correct Japanese alias.

Recommended content

I don’t read books because I’m lazy. Recently I have picked up the habit of reading light novels for the sake of learning Japanese (going through the Shinkanzen Master N2 Dokkai is boring).

YouTube Channels

I cannot stand having dinner without these. Also, shoutout to these channels that helped me score a 9.0 in the IELTS listening test (2023).

📺 Retrotech: This Does Not Compute, LGR, Action Retro, MichaelMJD

📚 Knowledge: Asianometry, ColdFusion, Johnny Harris (only for old videos)

Japanese learning materials

I don’t really follow any Japanese learning YouTube channels except Kaname Naito. I’ve always believed in learning the “right” way, by working through all the boring textbook stuff first. Once you’re solid enough to score around 150/180 on the JLPT N3 mock tests, that’s when real context-based learning starts to pay off.

For the beginner levels, just follow Minna no Nihongo. For N3, just follow the Mimikara Oboeru for vocabulary and Shinkanzen Master for everything else.

Click to expand: My take on immersion

That said, it doesn’t mean you should avoid native content altogether. I was lucky. During high school, I was completely obsessed with Love Live!. Its lyrics are surprisingly beginner-friendly (for the N5-N4 level) compared to most mainstream content made for native speakers. The lore behind it was that: the lyrics were supposed to be written by the characters, who were high school students, so they tended to be much simpler, definitely more approachable than something like YOASOBI. By the time I finished N4, I had already picked up a decent chunk of vocabulary that never showed up in textbooks like Minna no Nihongo.

The approach is simple: just listen to the songs while reading the lyrics. Over time, you’ll start to notice recurring vocabulary and grammar patterns. Just focus on picking those up.

Besides anime and light novels, I recommend checking out MariMariMari. It’s a channel made for native speakers (so the pace is very fast for a N3 holder), but it’s a great way to absorb natural language.

Log unfamiliar words into Anki. In addition to textbook-based decks, I keep a separate one just for this purpose: Vocab in the wild. But don’ overdo it. If you’re too aggressive, you’ll end up buried under a huge backlog of vocab in no time.

I think I will write a dedicated post on this topic later.