Two Sigma

The broad context here is I'm extremely happy with my job at Two Sigma. I've been here for 1.5 years, and in that time, I've learned and grown more than I would have thought possible.

• Almost everyone I've worked with in my corner of the Engineering world has been very kind, creative, and incredibly driven.
• Since team size is small relative to the number of products, impact-per-person can be extremely significant. It's possible to be a star, even as a new employee.
• Compensation is head-and-shoulders above similarly tenured roles in the rest of the industry.
• Casual, geeky, tech-focused culture relative to other firms in this sector.
• In-office extra-curriculars (giving talks, attending internal classes, organizing volunteering events) are readily available and encouraged by management.
• A strong lean towards the DIY in engineering projects means you'll probably learn a lot about parts of the software development process you might not elsewhere.
• A gold star on your resume. TS is a heavy hitter as far as perceived engineering clout.

On the flip side, growth can sometimes be painful.

• It somehow always seems like there's 20% more work to do than the amount of time there is to do it. Small teams with out-sized responsibilities, in combination with the type of passionate/ambitious personality that TS seems to attract, can produce an environment of intense self-imposed stress.
• That same strong DIY lean means a lot of projects get built from 0 to 1 by one or two people, and fail to make the leap from 1 to 2, as it's realized they're a lot more difficult to make stable and production-ready than it was to produce an MVP.
• As an adolescent company, TS is beginning to hit some phase changes, with things like maintenance of aging software, management of organizational structure, and knowledge sharing.
• Less hierarchical than many companies in its sector, but you can still feel like you're at the whims of a teetering stack of managerial decisions at times. It's often hard for bad news to propagate up the stack since people will soften what they say to higher-ups. This can lead to directional decisions being made without knowledge of the situation on the ground.
• Front-end development is largely unsupported by the build/test infrastructure and varies wildly from team to team, as each one learns separately to deal with the quirks.

Make stability, repeatability, and polish a first-order priority. Prefer to have one feature done excellently than three done as fast as possible.

Hire more onto existing initiatives relative to new ones. Work done by one person here might be done by an entire team elsewhere. It's great for learning new things, but it could burn people out.

Only add additional levels of management hierarchy if it's absolutely necessary to keep things from collapsing.