Yandex

I can't really imagine a better company in Russia, and I always have a very warm and nostalgic feeling when remembering the period when I worked at Yandex.

• Interesting and challenging problems with large impact.
• Very little pressure in most groups; there's rarely a formal deadline.
• This means you can often do a lot of experimentation in the direction you like; you will usually NOT be forced to implement a quick-and-dirty solution.
• Lots of super-smart people around.
• All the code is open to everybody in the company, so you can, for example, read some crazy stuff from the core search algorithms and learn a lot.
• Everybody is friends with everybody. It's a joy to come to work and just talk to people.
• A huge amount of intra-company communication infrastructure that's very actively used: mailing lists (with search), messenger, a wiki about EVERYTHING, "100 facts about everybody," people directory, etc. It really helps a lot to jump on board or just to creep around and learn more about who is who and what's going on in the company.
• Many of the mailing lists are cross-team (e.g., for Java developers in general) and they're active.
• Teams are very open. You can talk to anybody; you can send a suggestion to their mailing list and they'll respond, etc.
• Good infrastructure, for example, crazy stuff for load testing.
• The company is constantly collaborating with universities and helping host educational events. You can be proud to be part of a company that's having such a HUGE impact on education.
• Beautiful offices with on-site everything: food, sports equipment and instructors, English lessons, library, showers, games, etc.

Salary is middle of the market, not top.

There are fewer interesting tasks in St.-Petersburg than in Moscow (that's why I left).

Food in St.-Petersburg could be better, and if you take lunch after 3 p.m., you may find that all the tasty things are already eaten :)

There are, of course, some projects with smaller impact or less stellar management. You might be unlucky to run into one, then just try jumping to a different project.

There's a lot of "not invented here syndrome" all over the company. It can be dissatisfying to spend years developing something really cool and then see other people reimplementing a crappier version of it, just because it's fun for them too (with a formal reason like "it was hard to integrate"). Well, then learn better communication skills, mate, and don't concentrate just on coding.

Keep it up :)