HNHacker News
TopNewBestAskShowJobs

deepGem

2,284 karma · joined December 30, 2010

Generalist software guy, loves investing and writing code. Multiple battle scars in startups.

Email: pradeepbs at gmail dot com

submissionscomments
deepGem··on Productivity in 2017: analyzing 225 million hours of work time
Quite possible.
deepGem··on Productivity in 2017: analyzing 225 million hours of work time
My initial days of running had a dramatic positive effect on my productivity. However once my body got used to running I see no difference. I must admit I have stuck to the same pace and distance more or less. 5k in about 35 mins.
deepGem··on Clang runs in the browser and compiles C++ to WebAssembly
Thank you. In that case is it possible to access DOM elements within a C++ program the way we do in Javascript.
deepGem··on Clang runs in the browser and compiles C++ to WebAssembly
Sorry, for sounding this stupid. Does this imply you can now do web programming with C++ instead of Javascript.
deepGem··on Clang runs in the browser and compiles C++ to WebAssembly
This is really slick. Clang took about 20 secs to load on my quad core MBPro. How does this change web development ? Do we need an Ecmascript equivalent for C++ ?
deepGem··on Nepal bans solo climbers from Mount Everest
The racing analogy might have some flaws. Seat belts and helmets are for protection, they don't aid you in driving. Ice axes and supplemental oxygen aid you in climbing. You could very well climb without them. They are not safety devices. Seat belt is by definition a safety device which will save your life in case of a high speed accident. I think the only seat belt equivalent is the climbing rope.
deepGem··on The Long-Term Stock Exchange Comes to Life
I was initially very skeptical of this idea and I still am. Like many threads have pointed below, long term strategy is super easy to game. Oh 'we haven't made any money this quarter because we are debating and finalizing our long term goals'. This could become the anathema of rapid iteration.

NVidia and FB are perfect examples of consistent performance Q over Q. That doesn't mean they don't have a long term vision. I'm sure there are many such companies that I haven't heard of.

Companies not going public has little or nothing to do with the focus on short term earnings. Companies are not going public because 1. There are alternative sources of cheap cash PE funds etc. 2. The cost of doing an IPO is incredibly high (in terms of time). 3. Associated processes Sarbanes-Ox etc are a pain to manage.

CEOs are willing to sacrifice some equity to avoid all these hassles and remain private.

deepGem··on India’s IT industry laid off more than 56k employees this year
http://www.nasscom.in/sites/default/files/NASSCOM_Annual_Gui...

As per slide 2, IT outsourcing still is the largest contributor to the IT industry. It is not a subset. As per this report at least. I couldn't find any other credible sources.

"All engineers/engineering effort/R&D in country X is obviously substandard" This is obviously a very biased statement that suffers from the 'some/all' fallacy based on anecdotal evidences. Unfortunately the world works this way. What the majority sees and believes is what trumps, regardless of the truth.

deepGem··on Ask HN: What similarities might the crypto market have with the dot com bubble?
IMHO, most of the recent purchases of Bitcoin are mostly because of FOMO. That's the only explanation I can gather for something that increases in value so rapidly without any underlying change in the fundamentals. Quite a similar phenomenon happened during the dot com bubble too. Because of FOMO, people blindly invested in stocks that proved to be worthless later.

So the only similarity I see is irrational investment fueled by FOMO.

deepGem··on The Echo Dot was the best-selling product on Amazon this holiday season
I use it to play songs when guests come in and want to party. It's a huge pain to search for songs on Apple music and play them, when 10 guys are yelling their preferences. I also use the timer function quite a bit - for indoor exercises and cooking.
deepGem··on Magic Leap One
Some hedge funds do not have any uptight culture and they will most likely readily embrace the goofy glasses and gloves. Way better than a cluster of 6 monitors.
deepGem··on Litecoin and Ethereum buys and sells are temporarily disabled
What I meant was that if you distribute your money across all cryptocurrencies even if some of them become worthlesss you won’t have anything to regret, in he long run. I don’t know if anyone who had the foresight to cherrypick MSFT and AMZN. Most of them invested in the tech basket. This is just general diversification principle nothing new
deepGem··on Litecoin and Ethereum buys and sells are temporarily disabled
This is my belief too. I think if you can buy and hold for the next 5 years or so, massive volatility doesn't affect you. Volatility affects the day traders and the like.

To draw a parallel to the stock market is not appropriate, but those who bought into MSFT or AMZN even during the peak of the dot com bubble have nothing to regret today.

deepGem··on Office Parties Feel the Chill
Thanks, I'm trying to institutionalize this and was wondering what would be a good Christmas bonus. 250$ was my ballpark.
deepGem··on What Makes a Great Software Engineer? [pdf]
But don't the managers take the required estimates from engineers and plan accordingly ? Unless there is a strong deadline set by the market or by the customers.
deepGem··on What Makes a Great Software Engineer? [pdf]
I don't get the "9 to 5 wouldn't make you a great engineer"

As an experienced engineering manager this is what I've come across - Some of the engineers who put a 9 to 5 clause, are usually the obstinate engineers, who are not very productive during the hours of 9 to 5, but won't make up by working extra hours. The motto is 'This is all I'll deliver, live with it'. Perhaps, this bias factors in and a generalization emerges 'All 9 to 5 engineers are stubborn, non-creative, non-committed'. This is a huge fallacy because, again based on my limited experience, I have worked with some of the most productive, creative and committed engineers who mostly worked 9-5, but sprang to action in the middle of the night when a high priority production bug had to be looked into.

deepGem··on The Big Vitamin D Mistake
The NIH guys also did a study that said that even in tropical countries, Vitamin D absorption has fallen dramatically because of lifestyle changes. One of the probable factors is the lust for fair skin. I routinely see people covered from head to toe to help them stay fair skinned, even kids are sent to play covered in full sleeve clothing.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3897581/

deepGem··on Bitcoin ‘Ought to Be Outlawed,’ Nobel Prize Winner Stiglitz Says
And for people who wanted it to settle immediately or treat it like cash, they could make a direct transfer, , and have it clear on the order of minutes.

Can you please elaborate on this a bit more. What is direct transfer and how is it different from bitcoin /ethereum payments. From what I understand, they are the same.

deepGem··on Interactive workflows for C++ with Jupyter
Actually, I've found Xcode to be a very easily usable prototyping tool for C++. I haven't used it in any large project, but it's super simple to just prototype some concepts. Having that in a browser is even better.
deepGem··on Expensify sent images with personal data to Mechanical Turkers
Ok these were regular land images (I think at 10m resolution) which had some text descriptions on the images. We deemed the text to be confidential enough not to outsource. Def, not spy satellite stuff :)
deepGem··on Expensify sent images with personal data to Mechanical Turkers
I should clarify, we didn't use satellites directly, we just used the images given by a third party.
deepGem··on Bitcoin ‘Ought to Be Outlawed,’ Nobel Prize Winner Stiglitz Says
If we could run the economy on blockchain, why not?

One potential obstruction to this is probably the speed of transaction confirmation. I believe it takes about an hour for a transaction to be confirmed. If we equate that to a current fiat currency based transaction confirmation which takes just a fraction of a second, we are way off. I may be wrong in this comparison. What I do understand is that, you can't have the blockchain and also the speed of a fiat currency based transaction confirmation. What I don't understand is - Is this confirmation so crucial. I see blockchain as more amenable to existing workflows that takes eons to complete, such as an approval in an organization.

deepGem··on Expensify sent images with personal data to Mechanical Turkers
I'll be interested in understanding the technical safeguards as well, since I am breaking my head on how to do this. One of my projects was about doing OCR on some confidential satellite images and we didn't want to outsource the OCR process solely because we couldn't figure out a scalable way to obfuscate data selectively in images. One early attempt was to recognize numbers using MNIST, get the boundary and clip that boundary with a similar sized black square. This doesn't work for all forms of text though.
deepGem··on SpaceX raises another $100M
A CEO who promises the moon on a stick and delivers little is not doing his job well

But Elon is not promising the moon. He is very clear when he states his goals, that these goals are very ambitious goals and that the probability of meeting them is fairly low.

deepGem··on SpaceX raises another $100M
Unlike Tesla, Spacex is not a threat to any other publicly traded company AFAIK. So the barrage of negativity will be minimal. For every incompetent investor who comes on board, you'll have a great investor who'll trust you. This is the case in Tesla too. I mean, the odds that most of your investors will be incompetent is fairly low. It's not that much of a time sink, if you are running just one public company. Running two is a miracle. Jack Dorsey is doing it, so I wouldn't put it past Elon. All said and done, given a choice, Musk will keep Spacex private forever.
deepGem··on SpaceX raises another $100M
I am one of those guys. I don't really understand the value of pinterest. So a drop in performance made my exit far easier. I am probably not their target user.
deepGem··on Norvig's Python programs to practice or demonstrate skills
His Design of Computer programs course on Udacity is amazing.

https://in.udacity.com/course/design-of-computer-programs--c...

deepGem··on Business questions engineers should ask when interviewing at ML/AI companies
One of the primary factors that enabled Google to stay great was their ability to crawl and index the ever growing web at a rate no other company could.
deepGem··on Intel to Develop Discrete GPUs, Hires AMD's Raja Koduri as Chief Architect
True, I would love for some alternatives such as plaidml. However, I can't seem to fathom that Plaidml will be a worthy alternative to lets say Tensorflow or Pytorch or Caffe. I hope I am proven wrong.
deepGem··on Intel to Develop Discrete GPUs, Hires AMD's Raja Koduri as Chief Architect
This is what many people outside the AI world don’t seem to understand. Nvidia has a stranglehold in the form of CUDA and Cudnn. There isn’t any open source equivalent to Cudnn. AMD is trying to push OpenCl in this direction but it will be a long time before DL libraries start migrating to OpenCl. Like tomorrow by miracle if al alternative GPU which is as good as the 1080ti popped up, it would be useless in the AI market.
← PreviousPage 17 of 32Next →