Being a programmer, I usually advise people they do not want to get involved with custom software development. I suggest finding someone who is VERY knowledgeable with IB's applications and third party applications that integrate with IB and see if you can solve most of what you want with "off-the-shelf" solutions.
Thinking of hiring a programmer... Help very welcomed
- Thread starter xicaob
- Start date
First off, thanks for the great advice. Im total newbie as you can tell and is very much appreciated. Can you tell me an estimate for a bid for this kind of project? I was given a bid 2500 USD last week and now its seem like pennies when I read the comments you guys posted.
I will plan an alternative, less hired-labor consuming but I will love to have all the data for comparison.
Thanks RedDuke, as, same with MrMuppet, I would love to have an estimate of bid for this kind of project. I kindly ask you.
Thanks, I will look into it.
Thanks, Fractalize, and thats what Im looking for, control over my portfolio in the most optimized way possible (for me) and more free time. But It has to be cost effective, too, of course.
Thanks, mate!
All your comments are noted and taken into consideration. Thanks for your support.
Have a great one, and stay safe.
What you will get for that amount is probably quite quickly nailed together, "sort of works" quality and not something you want to rely on. Lowest programmers with IB API / trading knowledge offer is probably in the $40-50 per hour range. You can't build something decent in 50 hours.
I'd say at least 15k$ but I'd say 25k$ is more realistic for quality that keeps your sanity in check.
2500$ is more of a good programers wage for two weeks and I doubt that a project like yours is done in that short amount of time.
However:
If you can live with analytics only, things might be a little cheaper. If we're talking about sending, canceling, amending orders or conditional orders that work off of fill confirmations, things can get complicated really, really, (really) fast.
There are so many failsave mechanisms to consider, what your program will do in case of disconnections from the client (e.g. no service for your phone), how fast it needs to be, etc. etc.
I once outsourced an autospreader for markets that off the shelf software wasn't connected to and it was a really simple algorithm. Quote market 2 off of market 1 and if filled in market 2, hedge in market one.
To get this baby up and running took half a year of swearing and losing money trading because there were a couple of bugs that were really hard to find. And when you work sub ms speeds, you can blow up your account faster than you can blink.
Do NOT underestimate this.sss
2500$ is more of a good programers wage for two weeks and I doubt that a project like yours is done in that short amount of time.
However:
If you can live with analytics only, things might be a little cheaper. If we're talking about sending, canceling, amending orders or conditional orders that work off of fill confirmations, things can get complicated really, really, (really) fast.
There are so many failsave mechanisms to consider, what your program will do in case of disconnections from the client (e.g. no service for your phone), how fast it needs to be, etc. etc.
I once outsourced an autospreader for markets that off the shelf software wasn't connected to and it was a really simple algorithm. Quote market 2 off of market 1 and if filled in market 2, hedge in market one.
To get this baby up and running took half a year of swearing and losing money trading because there were a couple of bugs that were really hard to find. And when you work sub ms speeds, you can blow up your account faster than you can blink.
Do NOT underestimate this.sss
15K is cheap for half year engagement of a programmer! Where did you hire someone like that? I will want to hire too
Every non-developer underestimates the effort to get this right. The code part will not look like much but you really need to understand all aspects of it. Connections break between the gateway and IB servers, between the API client and gateway, then there are the connection issues and latency problems, what about when the data server disconnects and everything else is in order, what about having no confirmation of order submission etc.
If the developer hasn't written something autonomous before for IB API, I doubt he can get it right on the first 2-3 tries.
Spot on. It took us 2 years to iron out most bugs in our algos. That is why I asked how much money he manages. if it is not in 10s of millions, the uplift costs to get it properly working will be too big, and standard solution mix is the best way.
Plus the question about cost would requires detailed analysis, this is what non developers do not understand.
A couple of thoughts to add:
1. If you do this you're tied to IB, not matter how much they suck or how much they screw you. The sunk cost fallacy will kick your ass on this one, ask me how I know! So I highly recommend you do a search here for every IB complaint and make sure you're reasonably sure you'll never trade any of the products that people have identified as problematic. There are more than a couple.
2. I'll echo everyone here that you get what you pay for with developers. Coming from someone who runs a software intensive company and has run the whole gamut from Upwork sourced to personally sourced nearshore to locally sourced contractors to in-house employees. If you're not paying north of $75/hour for a sub 1-year contract you're getting crap and seeing false economy. The time to debug and fix something is multiples of the time to build it in the first place, double that again if it's someone new you have to bring in to debug a crappy persons code. That ignores the cost to you when your application does something it shouldn't or doesn't do something it should and directly costs you money.
3. If you do use Upwork be prepared to be charged 3-5X the number of hours as a good developer would charge you for the same product. I don't know if they suck and are actually using those hours or are padding their numbers. Be prepared for any change to cost way more than the original. Be prepared to be ghosted for weeks at a time when a more lucrative project comes along, then they act like nothing happened when they finally get back in touch with you. And that's from the "good" upwork developers.
Bottom line is that Google is hiring entry level developers for $100,000K per year. These are guys who aren't left alone with the code and have to abide by style guides, have all their code checked by senior devs, get a lot of hand holding, and have a full test team. If you think you can pay the hourly equivalent of $25K or $50K or even $100K per year and think you're going to get someone who can independently do mission critical financial software work when that guy could get a job for multiples of that if he was any good.....well let us know how it goes.
1. If you do this you're tied to IB, not matter how much they suck or how much they screw you. The sunk cost fallacy will kick your ass on this one, ask me how I know! So I highly recommend you do a search here for every IB complaint and make sure you're reasonably sure you'll never trade any of the products that people have identified as problematic. There are more than a couple.
2. I'll echo everyone here that you get what you pay for with developers. Coming from someone who runs a software intensive company and has run the whole gamut from Upwork sourced to personally sourced nearshore to locally sourced contractors to in-house employees. If you're not paying north of $75/hour for a sub 1-year contract you're getting crap and seeing false economy. The time to debug and fix something is multiples of the time to build it in the first place, double that again if it's someone new you have to bring in to debug a crappy persons code. That ignores the cost to you when your application does something it shouldn't or doesn't do something it should and directly costs you money.
3. If you do use Upwork be prepared to be charged 3-5X the number of hours as a good developer would charge you for the same product. I don't know if they suck and are actually using those hours or are padding their numbers. Be prepared for any change to cost way more than the original. Be prepared to be ghosted for weeks at a time when a more lucrative project comes along, then they act like nothing happened when they finally get back in touch with you. And that's from the "good" upwork developers.
Bottom line is that Google is hiring entry level developers for $100,000K per year. These are guys who aren't left alone with the code and have to abide by style guides, have all their code checked by senior devs, get a lot of hand holding, and have a full test team. If you think you can pay the hourly equivalent of $25K or $50K or even $100K per year and think you're going to get someone who can independently do mission critical financial software work when that guy could get a job for multiples of that if he was any good.....well let us know how it goes.