Ask HN: Transitioning from Industrial Research to Starting a Company | Hacker News Reader

Ask HN: Transitioning from Industrial Research to Starting a Company

7 pointsjiyunakenkyusha1 comment

Hello. I’m a computer science researcher working in an industrial lab. What made me determine to pursue a research career was feeling inspired during my high school and undergraduate years by researchers at Bell Labs and Xerox PARC and their work. I wanted to follow in their footsteps. However, the pressures of “publish and perish” are real, and moreover I want the freedom to set my own research direction. I have come to the conclusion that I won’t find the research environment that I want in either industry or academia, and that I’ll need to do my research on my own time. My research interests don’t require expensive equipment, just a computer and access to publications. The challenge is how to make the time, since I’m not independently wealthy and it’s difficult balancing a side project with a 40+ hr work week. It’s also very difficult to find part-time software engineering work.

One idea that has been on my mind is to find start some type of business that allows me to work 10-20 hours per week, giving me more time to pursue my research ideas independently without having to answer to management or tenure boards. Ideas that I’ve been thinking of include freelance software development, consulting, writing books, and selling software. However, I’m a bit at a loss regarding where to begin with this whole process, and I’d like to evaluate my options, as well as learn about options I haven’t considered. I’m not looking to become a tech billionaire; I want the freedom that comes with being my own boss.

You may or may not want to make a living off of your own technology.

You could have a research company that has a technology offering that pays the bills.

Alternatively, you could have a technology product company that does some research.

Depending on the situation, you may even be better off funding your research by selling something like real estate or cars without consideration for promoting or releasing your technology.

Regardless, you will need to strongly address at least one thing that clients or customers are ready to willingly pay for, as the foundation of your business.

Even if you operate as a non-profit organization you will still need a firm financial foundation to get going, and then perform research from there according to the resources you can rely on as they fluctuate through time.

And some things you may have to offer would have some clients who are more willing to purchase than others, and that can give you an idea how persuasive your sales efforts need to be. I wouldn't imagine putting less than 10 - 20 hours a week on sales or promotional efforts alone and that's also the amount of time I would never want research/experimental efforts to drop below. It's a constant balancing act and your outreach needs to be concentrated during regular working hours so you really have to focus on those business needs first. Then fit in the research pretty much the rest of the time, and it may be best to allow for some 24/7 operation to include nights, holidays, and weekends (when other "competitive"[0] researchers may not be making progress at all) before stabilizing at an acceptable work-life balance which is bound to be different than the mainstream no matter what. This was more apparent in chemical research which did require expensive equipment, and you want to get more done before the next calibration, maintenance or replacement cycle. The same cycles really that could be equivalent show-stoppers in computer science even if you have only a laptop when you think about it, you still have to pull the electronics in for a pit stop on a regular basis.

But to start a company you have to compromise and it may not even be possible without halting all research efforts entirely and instead concentrate on incorporating one or more established milestones or breakthroughs you already have under your belt.

Then sink or swim as a business before you can ramp up ongoing research again.

If successful, from that point you can possibly control the experiments, and own your own inventions.

[0] It may not really be competitive but you have built a business and that attracts wolves of many kinds to the door which you need to be prepared for. Competition for survival.

Reply on news.ycombinator.com