Ask HN: Complex tech idea with no coding experience - what to do?

3 points by stocktradr 12 years ago | 9 comments

As my name suggests, I do a lot of stock trading. In my past experience, I've ran website development divisions and what not. I have a lot of tech experience but minimal coding experience.

I have an idea uses a lot of principles from stock trading and brings it to the consumer front without them ever knowing. Vague description, yes.

My questions is how do you go about developing an idea with not much money and little coding experience? From my knowledge, the formulas I need are extremely complex and the coding required for them is complicated as well. I would try and learn but I can't seem to wrap my head around just javascript (and I know I need more than that).

I've looked up a lot of articles on this but every time I try to find someone to help, they require a ton of money or just aren't interested in putting in the time.

Any advice/ideas? Cheers!

carbocation 12 years ago |

Three options, with some color for each. In essence, this could be viewed as a time/money trade-off, although I think there's more to it than that:

(1) Learn to code. Start with a language that makes it easy to do what you need for the most complicated part of your app. What is that part? Do you use a research-grade algorithm that is only implemented in one particular language? Use that language. Do you use run-of-the-mill machine learning? Consider Python, or any language you like and just farm out the hard part to R. Do you just need to express complex algorithms without re-deriving them? Then really any language should do.

(2) Find a co-founder. This is tough because without any programming background, how do you evaluate whether someone would be a sufficiently technical co-founder for you? I'd say that this implies some aspect of #1 in that you'd need to get a bit more comfortable with programming.

(3) Outsource. This requires more money, less time. One might imagine that someone who does a lot of stock trading should actually have more money and less time than most, but I don't know you and it doesn't really matter anyway. You could outsource your technology. You'd just need to tightly control the schedule and the key elements of the product (e.g., make sure that they've correctly implemented your desired algorithms).

stocktradr 12 years ago | |

Cheers, thanks for the advice. Learning to program is definitely a barrier since much of the formulas I'm looking at using would require bit of power (makes me lean towards Python or Javascript - depending on end use).

I'm pretty leery about outsourcing. Back when I managed a web division, it always seemed that when we outsourced, things would go horribly wrong. I'll have to reach out to friends and find some good outsource recommendations.

Thanks for the advice!

kremdela 12 years ago |

Programmers perform the same type of risk assessment that you would do managing a portfolio. A coder is exchanging their time in exchange for your money or equity. Without a doubt, there is a challenge in finding someone with the right combination of experience, availability and interest in your project. If you are trying to attract coders, try to appeal to them my minimizing their risk. An experienced coder but first time entrepreneur will be more likely to join you if you can demonstrate a history of successes, experience executing / bringing products to market, or validation that there is a market for your product.

stocktradr 12 years ago | |

Great idea, thanks a ton. Thinking on it, it may be that my pitch to them is too complicated or convoluted. Its not their realm to know stocks (typically) so creating the pitch to peak their interest is a good starting point. Thanks again!

edent 12 years ago |

It's easy to have an idea - but it's hard to execute. Here's what I would do.

Step one - get an idea of the scale of the project. Usually the best thing to do is to write a description of every interaction the user will take with your service. Or, if you're able to, draw out the screens.

As a thought experiment, try doing this for a piece of software you already know. Think about all the different ways it can be used.

Ok, done?

Step two - what is the minimum thing you need to do in order to see if this idea is possible? The bare minimum.

Break it down into a series of very small steps.

For example, "The app needs to retrieve all tweets with stock symbol ___". Then, "The app needs to look up the stock price of symbol ___" Then, "If the tweet contains the word "awesome" and the stock price is over $X, sell the stock" etc.

Here's where you have to make a decision. Either learn to code each the bare minimum aspect, or find someone who can do it for you.

Having worked out exactly what you want your app to do, how many individual steps it will take to do it, and how the app should react in certain situations will make it a lot easier when you come to code it or hire someone to help.

So, refine your idea, then work on building it.

stocktradr 12 years ago | |

Great advice. Its all in my head at this point so this seems like the logical next step. I think this will also help me refine the pitch as well. Thanks a bunch!

kremdela 12 years ago | | |

No problem. Lean Startup (http://theleanstartup.com/principles) and Noah Kagan both have some great guides about methods to potentially validate your idea or find potential customers.

jehna1 12 years ago |

"brings it to the consumer"

If this a consumer product and your money is tight, you definitely want to start by making a crowdfunding project out of it (like Kickstarter or Indiegogo).

There you can raise the money and get your first critical group just by creating an awesome video out of your idea.

bowerbird 12 years ago |

are you rich?

if not, i don't trust your stock-trading advice.

if so, then why don't you believe in your idea enough to spend money to bring it to fruition?

if your idea requires "complicated coding" of formulas that are "extremely complex", then it's not surprising it will require "a ton of money".

and since you said you have run website development divisions "and what not", i'd think you'd know that.

-bowerbird