Are you asking because you're doing market research? If so I would strongly urge you to reconsider. If you're not doing market research you can probably disrega

Are you asking because you're doing market research? If so I would strongly urge you to reconsider. If you're not doing market research you can probably disregard most of this comment. Here are the realities of trading algorithms and trading automation: 1. Approximately no one trades competently - let alone profitably - outside of a trading firm. With very few exceptions the process of designing, implementing, testing, deploying, monitoring and improving trading algorithms is not a feasible job for a single person. This is aside from the problem of data acquisition, engineering and processing. There are people who profitably trade on their own, but they're typically veterans of a trading firm looking to grab a larger piece of the PnL on their own book. I don't know why there is this persistent belief that a lone individual is going to be automating trading competently - in the past I've compared this to thinking you could singlehandedly develop and/or maintain half of DuckDuckGo. 2. The firms which automate most of their trading - and this is increasingly the default - know how to do it and know how to acquire the talent requisite for maintaining that core competency. They like to keep things in-house where possible. They are extraordinarily secretive and don't like depending on third party infrastructure or tooling unless it's absolutely necessary. You are probably not going to develop an ultra-sticky product like Bloomberg, Nanex or kdb. A third party dependency that inserts itself into the risk management, analysis or execution pipeline is a liability. Moreover the intense secrecy means you will have difficulty talking to, or hearing from, employees of actual firms. 3. This is an incredibly crowded space to try and enter. Many companies have popped up which market products to firms looking to automate some or all of their process. They're typically not very successful in this effort for two reasons. The first is what I already mentioned about secrecy among firms. But just as importantly it's very difficult to develop a reliable trading process by hand, let alone to automate it. The customers you want to work with are savvy and most of the existing products are crap, so you'd need to have a very strong product and a very compelling case for why yours is different and useful. To put this all together for you: by asking this question on Hacker News you're strongly selecting for answers from people who almost certainly do not trade professionally. If they believe they are competently and effectively trading on their own without professional experience, it is most likely because they have vast unknown unknowns and a limited definition of what "automated trading" means (for example, they might think they're doing automated trading because they use the Interactive Brokers API instead of manually placing orders). You can target that kind of crowd, but if you do so you're entering the B2C cottage industry of (de facto) day trading education, not the B2B industry of trading itself. Unless you're attempting to build a fintech hypergrowth startup like Robinhood you really don't want to do this in a B2C capacity; then you're optimizing for customers who have both limited money and limited experience they could use to get any more money.