HNHacker News
TopNewBestAskShowJobs

robterrin

291 karma · joined July 27, 2015

www.robterrin.com www.tailrisk.com robert.terrin@tailrisk.com
submissionscomments
robterrin··on Probabilistic programming from scratch
Cool! Thanks for sharing.

I would also like to add, Stan is not that hard. Conceptually it's a bit of a jump, especially if you're coming from comp sci not statistics, but I think it's really worth it. Here are some helpful links:

Stan website: http://mc-stan.org/

A primer on Stan: http://www.stat.columbia.edu/~gelman/research/published/stan...

The Python interface PyStan documentation: https://pystan.readthedocs.io/en/latest/

A case using Stan in a real startup's product: https://www.smartly.io/blog/tutorial-how-we-productized-baye...

robterrin··on Ask HN: Your best advice when scouting customer dev interviewees.
Great question! I'm looking forward to hearing the answers. I think this is the hardest problem in entrepreneurship and why social capital and access are actually way more valuable than capital, technology, intelligence etc.

I'm in a similar situation, and my experience has been that there are three things that get you a busy person's time:

1. Provide immediate value. I run a cybersecurity consulting firm (www.tailriskconsulting.com) and execs like to learn a bit about the space, so the more I can do to offer actual knowledge right off the bat, the more likely they are to pick up the phone and let down their guard.

2. A personal connection. This could be a school you both attended, a personal hobby or activity you are both passionate about or another social bond. Join clubs, play sports, be open and try to learn from everybody you meet. Also, don't be afraid to self promote a bit. You have a brand whether you like it or not, it's up to you to shape it. Become the cybersecurity gal at your Brazilian Jiu Jitsu gym. Be the big data guy at your art classes. Talk about your stuff and be genuinely excited!

3. Warm intros. Your friends are likely to have more contacts than you do (https://en.wikipedia.org/wiki/Friendship_paradox). There are likely a few super connectors in your social circle. Find out who they are and try to tap in to that.

Finally, and this applies to all three approaches, work up the chain. Don't turn down coffee or a call with lower level folks; They can pass you up the chain. Don't give up! Rejection blows. We're all too self conscious too much of the time. If you believe what you're doing is important for the world, keep on plugging.

Best of luck! Let me know if there's anything I can do.

robterrin··on Before Silicon Valley, New Jersey Reigned As Nation's Center Of Innovation
And before NJ the US tech hub was Baltimore. Information technology has been built into the economic history of the United States for quite some time.

The first electronic telegraph ran from DC to Bmore. Prior to the telegraph and it's forerunner, the optical telegraph, the fastest way to transmit goods and information were the Clipper ships coming out of the port of Baltimore (https://en.wikipedia.org/wiki/Baltimore_Clipper).

If you're interested in reading more economic history I highly recommend both The Weather Experiment (https://www.amazon.com/Weather-Experiment-Pioneers-Sought-Fu...) and Empire of Cotton (https://www.amazon.com/Empire-Cotton-History-Sven-Beckert/dp...)

robterrin··on Launch HN: WittyThumbs (YC S17) – We help improve online-dating conversations
Thank you for creating something and trying to help people in one of life's most important activities. I want you to know that everything after this is coming from a place of genuine respect and support. My preemptive apologies for being blunt.

"The two of us had that sinking feeling in our early 20s that we’d missed our best opportunity to date," is a bad sign.

First, you were in your early 20's! Grow, learn, break a heart, get your heart broken, travel, read, get fired, start a nonprofit, experience life. Why were you so caught up in finding your optimal date while some of the best (i.e. most painful) growth opportunities were passing you by? I am not sure you have the perspective or life experience to be counseling people on something like dating.

Second, dating is not a game you can "win." One of my favorite professors ever, a statistics genius with possibly a touch of Aspergers, explained to our class one time how dating is like science. A "failed" experiment is knowledge gained about the world! A "failed" relationship is a chance to learn about yourself, others and relationships. The only way is to try.

Third, this is creating an arms race of banter instead of allowing people to genuinely connect. There's a famous OK Cupid article demonstrating why being your most authentic self is actually beneficial to your dating life (https://theblog.okcupid.com/the-mathematics-of-beauty-51bd25...).

There are a number of other reasons I think this is not a good way to spend your time, as a business, as a life pursuit and as a thing that detracts from the world, but I'll stop here so I don't sound like a complete jerk. I would strongly urge you to reconsider. Mostly, I think you just need time. Again, kudos on making a thing.

robterrin··on How Bayesian Inference Works (2016)
There are lots of discussions and explanations of what it means to be "Bayesian," but I think the best thing to do is jump in and start building models. That is how I came to understand the utility of Bayes.

If you're looking for a place to start I'd go to Andrew Gelman's introduction for the Stan Language: https://www.youtube.com/watch?v=T1gYvX5c2sM

There are Stan implementations in R, Python, Julia or you can run it in C++ since it's written in C++. I think this has greater potential to change how we deal with the unknown than AI or other machine learning.

robterrin··on OpenCrypto: Unchaining the JavaCard Ecosystem
I used to implement smart card as Personal Identity Verification for Federal employees. It was required by a regulation driven by the post 9/11 Bush era called HSPD-12 (Homeland Security Presidential Directive - 12).

It was a real boon to the DC area technology consulting business. Timed well with stimulus in the post crash crisis as well. Lots of investment in this technology (reminiscent of the Secret History of Silicon Valley) and surprised it has taken this long to get to finance and other uses.

I think one problem is smart cards were timed with the take off of mobile. They weren't very mobile friendly and seemed kind of redundant, but I prefer them to a fingerprint scanner on the iphone or galaxy.

Excited to see where this goes.

robterrin··on Empirical Bayes for multiple sample sizes
Stan is great! Glad to see it on HN. Nice write up too.
robterrin··on Rents in Megacities Can't Go Up Forever
I've heard this pitch for Chicago many times, and it seems reasonable if you are in the one of the industries, Tyler mentions, that are more diffuse.

The main problem with Chicago is it lacks an anchor industry. Is it the best place to be for finance? No, that's NYC. Is it healthcare? Boston, Baltimore, Cleveland or Rochester. Energy? Houston. Technology? SF. Entertainment? LA or NYC. Politics? DC.

What exactly is Chicago good at? They have Wilco, so that's cool.

robterrin··on Put Yourself on Internet with Jekyll, GitHub and Cloudflare
You might want to try Hexo (https://hexo.io/) or Wintersmith (http://wintersmith.io/). I use Hexo for my personal website and have really enjoyed it.

I tried Jekyll before, but between Ruby, some bugs and template choice overload I dropped it. Moved to Hexo a few months ago and built a couple projects with it. It's worked great for writing but also building a reviews website.

robterrin··on You Suck at Excel with Joel Spolsky (2015) [video]
Does anybody else feel that the vast majority of these things are either available in Google sheets or worth just doing in R? My last consulting project was on a VaR model built in VBA/SQL with a bunch of Excel sprinkled about and it was horrendous. Not Excel, VBA or SQL's fault, but it felt like a more thoughtful architect would have used better tools.
robterrin··on The CIA’s Venture-Capital Firm, Like Its Sponsor, Operates in the Shadows
Wow. I started my career in Identity Management and this is brilliant. Thank you for sharing.
robterrin··on Former Sequoia partners: The Midwest is the future of startups
Finance and technical people both underrate the power of tribalism, a major force in making any area a hub for a particular industry.
robterrin··on Index Innovation
Feedback and discussion would be much appreciated! Thanks.
robterrin··on I'm bullish on hedge funds
+1. Colloquially know as, "talking his own book."
robterrin··on Ask HN: Why don't companies hire programmers for fewer hours per day?
There's some truth and wisdom in your question and also, the replies that people have given. Likely, there is no one answer. A short incomplete list might look like this:

- coordination problem - organizational overhead - micromanagement

David Graeber wrote a good piece on bullshit jobs that you might enjoy: http://strikemag.org/bullshit-jobs/

robterrin··on The Superbook: Turn your smartphone into a laptop for $99
Palm built a mobile device 7 years before the iphone...
robterrin··on Where on Earth Are Threat Actors?
My first post here on Hacker News. Feedback appreciated. Thanks!
robterrin··on NYC tech catching up to Silicon Valley, while Seattle flails
Matt Zito hit the nail on the head. Let's not forget that the NYC finance workforce probably has about as many tech people as all of SF tech. GS has more programmers than FB after all. http://www.businessinsider.com/goldman-sachs-has-more-engine...

NYC definitely has a "money up front" attitude, but the enterprise ecosystem is really taking off here in part because there are so many customers, but also because of the great community builders like http://www.work-bench.com/

← PreviousPage 3 of 3