Hacker Newsnew | past | comments | ask | show | jobs | submit | khurram's commentslogin

When I worked in markets I would often get asked questions in the form of: should I short X? Should I trade X currency now or later?

More specifically many questioned whether they should buy GBP/USD the day before brexit or the day after. The only good advice is to buy 50% the day before and 50% the day after, that way you are wrong both times.


(worked at Goldman)

Despite having this, all the traders just end up using Bloomberg chat. GS tooling often gets in your way more than it helps.


By being one of the first to stand up you are able to get your luggage faster (if any), skip a few rows down the aisle, get out the plane faster, and ultimately get to the immigration line before everyone else. When flying to the US from the UK I would notice that this could sometimes save 1 hour+.

The end goal is to get out of the airport as fast as possible, not just the plane. So the advantage is pretty significant.


> People literally spend all their time doing this, if there was free money to be made someone would be making it:)

I used to think this, then I worked on a trading system in an investment bank. Don't underestimate how quickly and easily you can learn and exceed people who should know what they're doing, given sufficient motivation.


There are thousands of buy side firms all over the world stocked with stone cold geniuses who do this all day every day after having trained in maths and science all their lives.

The fact that you saw some people you perceived to be clowns once, does not demonstrate the market is anyone's for the taking.


Those people are usually managing a lot of money, though. Warren Buffet has said he has to trade completely differently nowadays due to size. So those people are not working the same opportunities most people here are going for.


A lot of HFT shops (if not most), don't "manage" any money in the conventional sense. They close the books at night and don't hold any positions when there isn't active trading.

What Buffet does and what HFT do are completely different. The reason why Buffet has to change how he trades is because of HFT. If Buffet puts in an order for 100,000 shares of Coke (KO) while it's at $40 a share, he would not stand a chance. HFT bots would swarm in and scoop every last share of KO that's available and then sell it back for incrementally more. All within a fraction of fraction of a fraction of a second.


Sure, even small hedge funds and propriety trading firms are generally working with millions of dollars.

I agree there must be some niches are that are amenable to exploitation by small independent traders (perhaps anything to do with nanocap stocks).

I have not had success finding niches where I thought I would be successful, but I could very well be insufficiently clever.

But trading highly liquid equities on a multi-day timescale (the strategy proposed here) is certainly not prohibitively hostile territory for large, sophisticated buy side operations.


So true. When you have a small account you can be much more agile and make some good returns simply by being able to trade without moving prices...


Wow! I wish I was in SF. Hopefully someone as friendly as you will be doing this in London (UK) next year.


Brilliant idea. I think this is pretty awesome.

However, from my (brief) usage of the landing page, I don't quite understand who will be doing the delegated work. Will it be some intern, or will it be your team? Who is your team? More specifically, what are there skills and what guarantees do I have?

Also - I can see potential problems with undergrad students submitting assignments to be done. Any idea how to regulate that kind of stuff and ensure you aren't just helping someone do something that they aren't allowed to delegate (or is this not your problem?).

I'm sure the details will be clearer with time, and for simple tasks this could work quite well.


Thank you! This gives us so much motivation.

As a group (and previous software engineering students) we talked about this. We define that, as soon as we detect that a given task is homework or project related, we wont do it. We prefer to nudge the student in the correct direction instead of receiving money for the student nor learning.

We understand that, although this is morally cool, we wont make money out of this. This is a price we are willing to pay: no homework and no software projects to prevent students from learning.

(we will probably add this as a disclaimer. Also I know that sometimes is difficult to triage a homework assignment out of professional development. This is where the human triage we are doing will help)

Thank you for bringing this up and for motivating us!


Clickable link: http://tweetsvu.com


Thanks for this.

I feel I can now go back and cover all the math i've forgotten since the first couple CS years (only a couple years ago!). I'm not really sure if it is worth the time from a practical standpoint, however I always feel slightly slower than other students when it comes to algorithm-related classes.

Hopefully this should get me up to speed again!


This sounds awesome. Unfortunately I don't want to login with my Facebook account.


I'm not sure what to think of this.

I can appreciate why you think this is a problem that needs to be solved. But I can't help feel a bit wrong about it.

From a technical standpoint also, I feel it is heavily flawed: So I write some sentimental notes, forget to check in one day or change my email address or some other very trivial reason not to get the notifications and boom, it thinks I'm dead and sends out messages that should not yet surface.

I think it's one of these things that is a 'good idea', but should never actually be followed through.


Thanks for the feedback.

I know what you mean. It's not quite what you'd expect from a web product these days. I actually had doubts about making this available to the public, at first. It was just a tool I wanted for myself, and a grim one at that.

Still, when it was done, there was no reason to keep it locked away. You never know. Writing all the copy took me longer than the actual coding, but I saw it as good training.

Anyway, this doesn't have to be as grim as the title suggests. It can be simply a kind of insurance: if you go on a trip and get lost, for instance, this thing can shoot important information to friends and family that helps bring you back. I could see this being relevant to people who live alone and travel a lot.

On the technical side, yes, the email warning system is weak by itself. That's why my first order of business was having it check your twitter feed. I just uploaded the feature.

In the future, it'll also check your facebook and foursquare activity. And send SMSes if you get really close to the timeout. Between social media monitoring, email and sms, it should actually be quite reliable at knowing whether or not you've disappeared.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: