HNHacker News
TopNewBestAskShowJobs

kkwteh

1,229 karma · joined May 9, 2013

Author of vinequai.com
submissionscomments
kkwteh··on Freeman Dyson Has Died
I particularly enjoyed being immersed in the ethos of engineering, which is very different from the ethos of science. A good scientist is a person with original ideas. A good engineer is a person who makes a design that works with as few original ideas as possible. There are no prima donnas in engineering.

Dyson said this while describing his time working on the Orion project. It had a profound effect on me and it has always stuck with me.

kkwteh··on Postdoc Myths
My guess is that it's mainly because postdocs are an unknown quantity without legible, interpretable experience, which greatly increases the risk of hiring them.

One of the main achievements of the Insight Data Science program (which I have participated in), is that it makes you produce output (a web app) that is legible to employers.

People are very risk-averse when making hiring decisions, and tend to hire people who know what they know.

kkwteh··on Tricks to start working despite not feeling like it
I use the small wins trick a lot. I start with the smallest win, which is writing down a small win on my todo list.

Then, when I see the small task, I feel motivated to do it just so I can have the pleasure of checking it off my list. By that time, I'm usually on a pretty good roll and keep working.

kkwteh··on Postdoc Myths
A lot of these problems postdocs are running seem to be the result of an inefficient allocation of talent by our current educational system.

All the smartest kids get tracked into getting math and physics PhDs and by the time they get to the postdoc stage it just becomes this insane impossible rat race tournament.

I saw the writing on the wall and got out of academia as soon as I defended my utterly mediocre PhD thesis in math.

I jumped on the data science wave in 2013 and all of a sudden everyone thought I was a genius for being able to run model.fit() and interpret the results.

These tracks that funnel you into cutthroat tournaments just seem like bad places to be.

Already data science is starting to become more of a commodity skill set.

Instead of trying to outcompete everyone I’m keeping my eyes open for my next big opportunity.

kkwteh··on Ask HN: What are the best unknown books you have read?
Reminds me a bit of Blood Meridian by Cormac McCarthy, which is an incredible book.
kkwteh··on What the Hell Was Modernism?
Reading this piece reaffirmed a lot of what I had been feeling in the SFMoma my last couple of visits. The place just felt lifeless. All the art was really disconnected from the world, and the abstract stuff seemed especially pointless.
kkwteh··on The Future of Mathematics? [video]
Looking at the code frequency, it looks like development of Lean 3 all but stopped around January 2018. The Lean 4 repo shows lots of activity since January 2019, but isn't in a useable form.

What are the future plans for the project? How will this be distributed in a form that mathematicians can use and contribute to?

kkwteh··on How can we develop transformative tools for thought?
Daniel Dennett introduced the introducing notion of Intuition Pump, which I would say definitely qualify as transformative tools for thought.

They are thought experiments designed to focus attention on important features of a question and to deflect them from getting bogged down in hard-to-follow details.

Google talk by Daniel Dennett on Intuition Pumps https://www.youtube.com/watch?v=4Q_mY54hjM0

kkwteh··on The Transformation of the Textbook
It would have been great to have digital textbooks in university. I had to walk a lot to get around campus, and I could only carry at the very most three textbooks at the time. Not having access to all of my books at all times was definitely an impediment to learning.
kkwteh··on We Need a New Science of Progress
I don’t know if it’s that far off. Our current notions of progress haven't allowed us to avoid our current condition of extreme indebtedness and financial insecurity and I think we're at a point where we need to reexamine these notions. I don't think the author does enough to do that.
kkwteh··on Ask HN: What people skills do you wish you learned earlier in your career?
I do agree that most people are boring if you don't try and figure out what's interesting about them.

Discovering what's interesting about someone is an art. The totality of people you will meet have a much wider variety of experience than you could ever accumulate on your own. When I meet someone new, the first thing I try to find out is what it is they have done that I have not, and then I ask them about that. Often, I learn something new and interesting.

In general, people will never realize on their own what you'll find interesting about them, so you have to seek it out actively.

kkwteh··on The U.S. Labor Market Isn’t All That Healthy
Unfiltered anecdotal comments are often much richer than the kind of piece that makes it through the editorial process of a large publication like Bloomberg. That's not to say I don't take it with a grain of salt.
kkwteh··on The U.S. Labor Market Isn’t All That Healthy
That 30% MBA unemployment figure is much more striking to me than the article. Any thoughts as to why your friends can't find jobs? Where were they planning on working?
kkwteh··on IQ rates are dropping in many developed countries
https://medium.com/incerto/iq-is-largely-a-pseudoscientific-...
kkwteh··on IQ rates are dropping in many developed countries
I agree with other replies that salary is a very flawed indicator of intelligence, but it was a simple data point to share. My only point is that it's not at all obvious that IQ is a valid measure of intelligence.

If you think about what is involved in an IQ test and all the ways someone's intelligence manifests itself, it seems clear to me that IQ is at best a very distorted and narrow measurement of intelligence.

kkwteh··on IQ rates are dropping in many developed countries
One huge assumption no one has challenged is the notion that IQ is a valid measure of intelligence. One reason to doubt this is that above a low threshold (80 IIRC), IQ has no correlation with salary.
kkwteh··on How random can you be?
I wrote a quick and dirty Python script that outputs a string of guesses that will beat the guesser. It keeps track of the 5-gram counters and always opts for the less-used option. If there is no less-used option it picks a random direction.

For instance, if you input RLLRRLLLLRLLRRRLLRLLLLLLRRLRLRLLLRRRRLRRRRRRLLLRLRLRRLRRLLRRLLLRRLRRLRLLRLRRRLRLRLLLRLLRLLLLLRLRLRRR it only guesses right 34% of the time.

https://gist.github.com/kkwteh/b81d8e599ec46a2d64b096f953a11...

kkwteh··on List of 30 European Tech Visas and Work Permits
I moved to Paris from San Francisco. The startup I joined uses English as its working language, so the language barrier hasn't been a problem at all. Also I found French to be relatively similar to English and quite easy to learn. It's been a huge joy being able to speak to people in a second language. I've wanted to learn a second language for a long time and I finally have a good reason to do it.

Also there are interesting startups in Paris outside of fintech, like Alan (where I work), Algolia, and Datadog. Airbnb, Facebook and Google have offices here as well I believe.

I've never been to Berlin, so I can't really compare, but I really enjoy living in Paris so far. I certainly like Paris more than San Francisco.

I wrote a blog post about my move a few weeks ago: https://medium.com/@kkwteh/why-i-moved-from-san-francisco-to...

kkwteh··on Why I Moved from San Francisco to Paris and joined Alan
Here's an old blog post with the salary grid for the company. It's a bit out of date. Today the salaries are somewhat higher and equity grants are smaller.

https://www.maddyness.com/2017/06/25/maddyrex-pourquoi-grill...

kkwteh··on Why I Moved from San Francisco to Paris and joined Alan
It's not too much of a barrier since everyone speaks English in the office. Generally everyone gets along pretty well in the office, English and French speakers alike.
kkwteh··on Why I Moved from San Francisco to Paris and joined Alan
Author here.

My salary in Paris is definitely quite a bit lower than what it was in San Francisco, but I think the lower cost of health care, child care, education and housing largely make up for it. In particular, I feel much less exposed in terms of negative tail events. I'm much less worried about a chronic medical condition or college to wipe me or my future family out financially.

Totally agree about language, healthcare, food, metro, and work-life balance. There are some things money can't buy.

kkwteh··on Why I Moved from San Francisco to Paris and joined Alan
I'm the writer.

I have no doubt that I wouldn't be treated as kindly if I weren't skilled and educated. However they do treat me pretty well here. In the US all of the immigration officials I have to interact with seem to be getting more and more hostile. Markedly so since Trump was elected.

kkwteh··on Ask HN: What books changed the way you think about almost everything?
Antifragile has changed the way I've thought about many things. A couple key concepts in the book are like lenses that gave me entirely new ways of viewing the world.

The Lindy effect: The strongest systems are usually the ones that have been around the longest. Accordingly, there should be a healthy distrust for any recent invention or practice.

The world is too complex to understand: Complex systems are mostly not designed, but exist and thrive by virtue of natural selection. This includes human societies. Even leaders and managers of these systems do not understand why they work. In particular you can't predict the outcome of tampering with complex systems, and you should avoid doing it in a wholesale centralized way.

Stress makes antifragile systems stronger: In particular a certain amount of stress, challenge, and variation is necessary for a person to remain strong.

Extreme tails: Fragile and antifragile systems often have payouts where a single event dominates the outcome. As an individual, there are two takeaways. There are only a couple of opportunities in life that count. They will only count if you maintain the optionality to take advantage of them. A single tragedy can ruin you. You have to look out for it constantly, even if everything has been pretty rosy up until now.

kkwteh··on Job Promotion is a Trap
Two hours reviewing one resumé is way too much. A few minutes should be plenty of time to decide whether or not to start the interview process with a candidate.
kkwteh··on Bullet Journaling (2016)
At work, I use elements of bullet journaling but instead of using a paper journal I use my personal DM channel in Slack. This way I can quickly paste useful code / data into my journal and run searches efficiently.
kkwteh··on Ask HN: How to start a startup that requires a high amount of initial capital?
You can raise a lot of money at the beginning if the founders have the right connections and track record. Opendoor was able to raise a $10M Series A without a product. [0]

[0] https://techcrunch.com/2014/07/07/opendoor/

kkwteh··on Comparing the NYT op-ed to tweets from cabinet members
The NYTimes stated just above the anonymous opinion that they know who the official is, so I think it's safe to say that it wasn't written by Trump.
kkwteh··on Ask HN: What's your hobby?
Argentine Tango. Despite what you might think, it's a surprisingly technical activity. I meet a surprisingly high number of engineers and PhDs in the tango community. When I'm dancing, it puts me into a state of flow where I'm constantly listening to the music and thinking which movements will fit best. If you're in the SF Bay Area, it's arguably the best city for tango in the US.
kkwteh··on 29 Design Features That Increase Your Home’s Value
Having the phrase "solar panel" in a property description may correlate with a 40% increase in "expected sale price" (the blog post doesn't say what they mean by this), but having solar panels in your home correlates with all sorts of things that haven't been controlled for in the analysis (for instance how warm the local climate is).

Much more useful would be an estimate of how much a home's value would increase if solar panels were installed. That figure is certainly much less than 40%.

Reference: I built Opendoor's home valuation model.

kkwteh··on Show HN: A little eye-candy for the terminal (in Go)
Looks like an off by one error. I'm seeing this error as well for some of the fonts.
← PreviousPage 3 of 4Next →