What to Look for in Startups

How Much Can I Learn?

Learning is, on some levels, the only goal. All else grows from getting better at life, the universe, and everything.

Don't be evil

Is the company's effect positive (e.g. kiva or goodguide). Don't be Farmville, where a majority of revenue comes from psychologically addicted users paying for in-game currency. 

Have Impact On the World

What's the end goal? Change the world or put change in your pocket?

Employees Can Drive, are Given Freedom

Early-stage companies still have most of the risk ahead of them. Early employees are quasi-founders and should be given a seat at the table. Empowered employees are happier and perform better. If the founders have 100x the equity of your first few employees, like most companies, then you're doing it wrong.

Internal Impact, Not Face-Time

Caring about hours-in-office instead of results is a sign of management ill-suited for engineering. Engineers make non-linear contributions.

Expertise Guides, Data Drives

While you need domain expertise to give your company direction, the day-to-day operations should be driven as much by data and science as possible. Startups are attempting entirely new things; Even experts have the wrong ideas all the time. Do you have an analytics dashboard where you can track the health of everything you care about at all times? Do you A/B test as much as you can?

Transparency

Internal transparency about company information. Transparency to your users for trust and long-term growth.  "Beware of he who would deny you access to information, for in his heart he dreams himself your master".


The Beer Test
Getting along well with your peers makes you happier, but it also helps honest dialogue over difficult subjects. And you can work with each other 2 weeks into crunch time at midnight.

Smart Workplace

Don't be penny-wise but pound-foolish with your tools and environment. If a keyboard, monitor(s), chair, and programming environment is how you get your work done, they need to enable.


Engineering Results, not Engineering Fashion.
It's good to be aware of them, but don't waste time constantly installing/rei-installing/configuring/researching the most-recent, fashionable framework unless it seriously helps you. Consider the Lindy Effect: The longer a technology has been around, the longer it will stay around; It has been battle-tested.


You can't have everything you want, though. An amazing work environment, great compensation, and great coworkers can overshadow a mission that's not exactly to your liking, and vice-versa.