carlmcqueen

For awhile I was working on Monte Carlo sims in my job in finance in my early career. Just re-building a existing archaic excel monster in python to be more flexible to new investment models and implement and allow for more levers. Since I was already working with them daily I begin applying Monte Carlo models to a lot more problems I was thinking about. It truly is a fun tool to play with, especially when you're in the thick of designing them.

Anecdotal, and old now: I worked with an ex-NSA agent when I worked at a big bank who worked out of a some of middle east offices in early 2000s. He talked about how new agents often struggle with the size of data (even then) but most good agents work immediately to look at the lack of normal data. Criminals/targets have their own signal of data and by filtering traditional data patterns you're left with a smaller dataset of the targets you're there to find. He used the same patterns to find financial white collar cheaters in bank data.

(example: phones off during day, on from 1am-5am then shut off again, no facebook browsing at all, etc.)

I'm not far into it, so I don't know how good it will be. However FX's 'Class of 2009' seems interesting too where FBI agents have AI strapped to them feeding the dataset at all times. The show jumps between 2009, 2023, and like 2035 or something and the implications of such a system.

Having to ask people be kind about your code on HN is such a shame of reality to me. Half of brilliance is having your working version, used by many people. You had working code that brought reddit to so many reddit users every day that was better than what the official app of reddit provided.

Some of the 'code' running massively profitable departments in finance are spreadsheets, on shared drives, with thousands of linked cells thought up by a brilliant employee who left the company five or six years in the past.

I wish I was kidding.

Thanks for how great you kept apollo.

I know it's my own niche case, but with apple silicone I have more issues with external monitors than I've ever had in my entire life. I run my own business and I've always liked Macs because having the same software as my team makes fixing issues so much quicker but this is the first time I've thought about moving us to PCs.

In my case, it is a start of the day, every single day, issue where I'm unplugging and re-plugging in monitors waiting for it to 'take' and then it's fine for the day.

The fear of language model AI takes me back to my original fear when I joined a data science team at a huge bank with just a bachelor's degree and five years of experience. I was embedded in the mess of databases created by hundreds of large bank mergers.

I was working with really, really smart PHDs and was doubting my hiring to the team. I felt like a lot of these guys had forgotten more than I'd ever known.

During my time in that job, I realized that my experience in navigating the complex databases from working with the business and operational teams gave me an advantage. I took my domain knowledge for granted and was able to work on the problems at hand much faster than the highly intelligent colleagues who relied on IT/DBAs to write their queries. They often had to go back and forth for a couple of days to clarify their requests, leading to misunderstandings and delays.

In the end, I fit in just fine, held my own, and was aware of my individual talents. I enjoyed learning from the PhDs, and they were happy to teach from their backgrounds.

A lot of the experience from the big bank has 0 transferability to any other company. What made me stand out was I knew which tables of the gigantic data environment were the best to use, I had hundreds of already built queries for many different problems and I was dependable. It got me really far before I went off on my own.

Personal response to this, despite owning a home, is that there always seems to be some reason the media/news stories are making the home market sound shaky when they're actually talking about the viability of the house flipping market. The same can be said for wall street with big reports of days that have large losses, etc when over time its generally a longer term win.

I will say the value of not having to worry about your toilets, oven, etc when you're career centered has intrinsic value not tied into the money you'd make over time owning the asset. I hate having to deal with that stuff with the little free time I have.

I live in the midwest and own a non-tesla EV, and the minute you get out of any kind of bigger cities your chances of finding a level 3 are quite sparse. There are a LOT of chargepoint level 2s, but level 3s seem to be at car dealerships who don't offer it to everyone, hotels that don't allow it for everyone, etc.

We did a 80 mile each way day trip to visit someone and bring them back and had to line up lunch in an area with no great food options so we could sit for the 30 minutes to get back up to 80%.

With it being cold in the midwest our advertised '280 range' or '180 range at 80%' never is even close to that.

It's like the olden days before really efficient ICE vehicles.

Really interesting point but I know I've gotten older when he said 'my audience is middle-aged' but the way I read the chart from google is that most of his audience is between 21-34 with a tail of older than that and a steep drop off after the age of 40. That is older millennials for sure, but is that 'middle aged'?

Middle aged traditionally is 40-60.

Oracle people tend to be heavy oracle people too.

At the bank I worked before 2019 at I was on a advanced analytic team, we had this amazing teradata database and then there were these insanely fast (yet older) IBM DB2 databases and then there was a few big oracle databases.

We did amazing things with teradata + DB2 and then a leader who was tired of multiple databases asked us to vote, we chose teradata so we went with oracle and the migration was so bad I left.

Hulu is one, if you need an example.

This is an analysis should probably be read while listening to Dave Matthews Band Ants Marching. Really cool analysis and fun to see it written in R.

Jerr Thorpe had an app that gave you your power back and tracked your phone info in a way you could have yourself for awhile but it became abandonware and started eating battery life so I took it off my phone. OpenPaths, that was it's name. https://www.jerthorp.com/openpaths

I'm not a doctor, but as a family member and extended family member, I know that many bipolar children appear with ADHD in their early years and are treated with ADHD medication.

They are treated with stimulants that have no effect on bipolar disorder.

I don't know why I never looked into it more but this comment caught me off guard. I always wondered what hotline was because a friend of mine in high school's dad was a bartender part time in the carribbean and during the summer in chicago and he would source things his son needed or wanted through 'hotline' and I could never find it as a windows user.

In 2008, I graduated at the same time as this guy, and I had two film grad friends who felt EXACTLY the same way.

One started as a wire guy in television and worked his way up to a producer now, the other one I got a job in corporate banking and he moonlit in a wedding video company. He has his own wedding video company now.

I wasted my college experience as an awkward introvert. I didn't take college to be a time to make connections with my professors, I didn't take the time to see it as a time of ecosystem of closeness you'd never have again.

I was way to shell-shocked inwards. The becoming an adult wasn't something I was ready for yet. I look back at all the opportunities I had to really learn interesting things and network and grow and frown because I was still in my high school 'just survive school' mentality. I did as little as I could to get through each class.

Why would only having been with three people change how much sex you have had?

If journalism is similar to art, this seems obvious. When we consider the many classical periods of art, we see that the majority of the artists lived in the same location, knew each other, and had the same circle of funders for their work. It is essential to be able to compete with someone on your level.

dtplyr is an option as well which lets you use tidyverse syntax with data.table backend. Speed and syntax.

If you learned data.table however, it's better to just stay in data.table. Nothing in the tidyverse can touch the efficiency of data table.

My wife has a doctorate in occupational therapy and works as a hand therapist, when pay is the only metric looked at you really do miss a lot.

The body condition of construction workers, fork lift operators and even welders is worn down and in pain.

Not to mention, they have to live with unfair medical standards. When she does hand strength assessments for workers comp the number is based on natural average. So lets say a normal office worker squeezes on the test at a score of 100, a construction worker squeezes at 320. Workers comp says they can return to work if they can squeeze at like 120. Which terrifies the construction workers but they won't get any more time off.

Or even continually change it. So even if you agreed with original terms, to continue using it you have to keep accepting the changes.

When I worked for a major financial institution in a data science / analytic group there were a lot of very, very smart math types who couldn't code. They'd ask how I'd gotten my code working and I'd never lie, that many parts of it I'd fixed using SO. This was 2012-2015 when a lot of the machine learning wasn't in helpful R packages or python libraries pre-written.

Many would ask to see the page I'd read and I'd show them and they would ask how I got my code working from an example that had nothing to do with what our task was.

I think there is a lot of value to looking at a SO answer, generalizing it and making it work in your code and doing it quickly. Knowing which SO articles are junk and which ones are gems is something many take for granted but is actually something close to having the right 'gut' feeling.

https://www.cdc.gov/tuskegee/timeline.htm

Which has been done before.

This is a very common answer to these stories on hackernews but this one is from a humble point of view that truly brings home the point.

My side is that I worked for a bank on the brokerage side for ten years in different positions. What always struck me was that my access was very carefully controlled, I was a background checked employee and had to meet with compliance once a year, etc etc.

However when a law firm asked for anything or consultants said they needed more data they just sent massive data dumps to the network admin guy, no questions further asked. At least not at my pay grade.

As I've consulted I ask for only what I need to keep my own risk down but it is always a surprise to my clients I don't want PII I don't need and only the data that my model will help enhance.

This is good news for new users to python when getting books that start with an argument for using 2 or 3 and then the book trying to be in both.

while software driven, there is a program you can get called Haptic Touch Bar, which vibrates when you hit the touchbar buttons.

Not perfect, but at least sort of helpful.

The main point, let me have a nice version without the touchbar is shared.

A lot of the discussions of how AI is coming for your jobs follows the same reasoning of how automation is/was, even before it was called machine learning.

The data science team at a major bank I used to work for historically had a team that sourced and cleaned the data, then a team who would explore and build the models and then a team who would learn and run the models going forward. In my opinion the final team was the most important and the most tragic for being condensed, they noticed when the models needed to return to the model builders to be adjusted when missing the mark.

The "lost jobs" in this case is that when I was on the team we had to learn the entire bank database structure and source our own data, build the models and then automate them in such ways to "catch" when they were missing the mark.

The team will be further shrunk as the software tools provided have better auto-sql for sourcing the data, automated model building functions, and automated visualizations thus removing even that as a special skills.

All this AI stuff is nice, but the important questions are what are the benchmarks of a retropie?

Not to spoil the article for anyone but..

Pretty in depth article and well laid out explanation of natural gradient descent with a small pre-fit dataset for a conclusion of 'too computationally expensive for machine learning/big data world'.

This is what I struggled with in school. You'd spend a class week learning some tough stuff only to be told 'this is no longer done, better methods are now used.'

Sometimes the work is needed to allow you to understand why/how the new method is used, but in many cases I didn't find that to be true.

I've only see this in parallels when using a windows application.

So if you have windows office because it has traditionally been better than the one offered to mac you'd have to remember to use control instead of command.

https://www.stltoday.com/news/local/crime-and-courts/st-loui...

Wasn't a homeless person, but in St. Louis Missouri a house isn't out of the question.