Yan (@yanoak) 's Twitter Profile
Yan

@yanoak

Burmese guy. Geek. Interested in #dataviz, #opendata, #civictech. Founder @thibitweets

ID: 19457301

calendar_today24-01-2009 18:38:07

21,21K Tweet

4,4K Followers

5,5K Following

Thibi.co (@thibi_co) 's Twitter Profile Photo

🧵 Land #reclamation projects have already left drastic impacts on coastal communities in the #Philippines. Explore how such developments are reshaping both the #environment and the lives of local residents in this interactive data story by AlterMidya and Thibi. ⌄

Taylor Baldwin (@taylorbaldwin) 's Twitter Profile Photo

Spent the last few weeks parsing NYC's massive 2017 lidar dataset (almost 1TB of data) and segmenting them by building. Now you can select any of NYC's more than 1M buildings and see its point cloud. #webgpu

Ethan Mollick (@emollick) 's Twitter Profile Photo

Written before the current wave of generative AI, this is an interesting historically-informed view of whether computers can create “Art.” It deals thoughtfully with the difference between AI as tool and AI as artist, and how definitions shift over time. arxiv.org/pdf/1801.04486

Written before the current wave of generative AI, this is an interesting historically-informed view of whether computers can create “Art.”

It deals thoughtfully with the difference between AI as tool and AI as artist, and how definitions shift over time. arxiv.org/pdf/1801.04486
Yan (@yanoak) 's Twitter Profile Photo

Digging up all this stuff I used to be really into about new work and new culture. This is the #Solarpunk ideal described in a manifesto before it was a meme newwork-newculture.dev/what/

Thibi.co (@thibi_co) 's Twitter Profile Photo

Thanks to fellow data analysts and web developers from Open Development Cambodia Organization who participated in our training session on #Scrollytelling Using Maps led by Yan Naung Oak, data trainer and founder of Thibi. ⌄

Thanks to fellow data analysts and web developers from <a href="/opendevcam/">Open Development Cambodia Organization</a> who participated in our training session on #Scrollytelling Using Maps led by Yan Naung Oak, data trainer and founder of Thibi. ⌄
MIMU (@the_mimu) 's Twitter Profile Photo

More flood maps (based on data as of 12 Sep 2024 morning) have been released in mixed formats: larger scale (more detail and village points) and smaller scale (overview with area and population estimates). bit.ly/MMFlood24 - Mon and Kayin States - Bago Region

More flood maps (based on data as of 12 Sep 2024 morning) have been released in mixed formats: larger scale (more detail and village points) and smaller scale (overview with area and population estimates). bit.ly/MMFlood24

- Mon and Kayin States
- Bago Region
Yan (@yanoak) 's Twitter Profile Photo

Been coding up 2 side projects and wanted to commit some proper time to it. So I'm starting a #100DaysOfCode challenge to #buildinpublic Project 1 is building a tool that lets you browse book recommendations from podcasts. Yesterday was Day 1. Got some transcripts via #whisper

Been coding up 2 side projects and wanted to commit some proper time to it. So I'm starting a #100DaysOfCode challenge to #buildinpublic

Project 1 is building a tool that lets you browse book recommendations from podcasts.

Yesterday was Day 1. Got some transcripts via #whisper
Ethan Mollick (@emollick) 's Twitter Profile Photo

The first author of this astrophysics paper found that if he gave o1-preview the methods section, it was able to reproduce 10 months of work coding he did as a PhD in 5 prompts (a few caveats in the video) Side note: all of your methods sections are becoming instruction manuals.

The first author of this astrophysics paper found that if he gave o1-preview the methods section, it was able to reproduce 10 months of work coding he did as a PhD in 5 prompts (a few caveats in the video)

Side note: all of your methods sections are becoming instruction manuals.
Yan (@yanoak) 's Twitter Profile Photo

Day 2 of #100DaysofCode building a tool that lets you browse book recommendations from podcasts. Quickly realising the I have to move the data processing pipeline to the cloud because I downloaded the audio for just one podcast and it's already 22.25 GB. #buildinpublic

Day 2 of #100DaysofCode building a tool that lets you browse book recommendations from podcasts.

Quickly realising the I have to move the data processing pipeline to the cloud because I downloaded the audio for just one podcast and it's already 22.25 GB.

#buildinpublic
Mogwai (@mogwaiband) 's Twitter Profile Photo

2/3. World Tour Announced, starting in 2025 with dates in Europe, Asia and North America ( more tba). Tickets on sale Fri 20 Sep @ 10am local time ➡️ mogwai.scot

2/3.

World Tour Announced, starting in 2025 with dates in Europe, Asia and North America ( more tba).

Tickets on sale Fri 20 Sep @ 10am local time ➡️ mogwai.scot
Yan (@yanoak) 's Twitter Profile Photo

Days 3 and 4 of #100DaysOfCode Got my script to run on EC2. Almost finished parsing all the transcripts for the first podcast. #BuildInPublic

Yan (@yanoak) 's Twitter Profile Photo

Day 5 of #100DaysOfCode My little EC2 instance churned out all the transcripts after about 30 hours. Then I piped the transcripts through GPT4o-mini and got the snippets of the podcast guests' recommendations for each book! All the data I need is almost ready. #BuildInPublic

Day 5 of #100DaysOfCode 

My little EC2 instance churned out all the transcripts after about 30 hours. Then I piped the transcripts through GPT4o-mini and got the snippets of the podcast guests' recommendations for each book!

All the data I need is almost ready. #BuildInPublic
Yan (@yanoak) 's Twitter Profile Photo

Day 6 of #100DaysOfCode Read and watched tutorials about RAGs and agents. My project doesn't need them yet, but soon I do want to feed entire podcast transcripts into an LLM to be made sense of, and not just extract the structured book recs. But first, to ship! #buildinpublic

Yan (@yanoak) 's Twitter Profile Photo

Tried out Bardeen for some simple web scraping, and I lost my patience quite quickly after a few tries. I think I'd have had more patience if I wasn't expecting it to just work automagically because *wooo* AI *wooo* Back to trusty old webscraper.io which just works

The Prepaid Economy (@prepaid_africa) 's Twitter Profile Photo

hawker culture wasn’t built on standardisation & economies of scale. It was built on individual craft & infinite diversity. also 👉🏽public-funded capital injection for micro-businesses + working-class families after 1965, when Singapore a sovereign state vittlesmagazine.com/p/the-life-and…

Yan (@yanoak) 's Twitter Profile Photo

Days 7 and 8 of #100DaysOfCode Got sidetracked and excited about working on another side project whilst working on my side project. It's a bit of web scraping and mapboxing. I recently got diagnosed with ADHD. The doctor was like, yes you obviously have it. #BuildInPublic

Yan (@yanoak) 's Twitter Profile Photo

Days 9 and 10 #100DaysOfCode OK all my data is prepped now. Time for a basic UI. Here's a literal napkin sketch of the first iteration of the UI. #BuildInPublic #sketchonnapkin

Days 9 and 10 #100DaysOfCode 

OK all my data is prepped now. Time for a basic UI. 

Here's a literal napkin sketch of the first iteration of the UI. #BuildInPublic #sketchonnapkin