Pros
IMHO Best IT employer in Russia Very high expertise, lots to learn from colleagues Wide variety of bleeding-edge projects: SDC, Search, Taxi, Cloud, Geoservices, Media etc. You can learn from all of them, and most of them affect from tens of thousands to many millions users. The scale is enormous, which makes a very unique experience. The external impact is giant. Yandex is very prestigious in Russia, and generally it is a "good company" (Google's "don't be evil") Wonderful office in the center. Open-minded people, bright minds around, nice corporate culture, not official atmosphere. Equality, easy communication with higher level management and other teams. Working hours are flexible and not controlled -- you just have to deliver the results. Medical insurance. Meals compensation. Many cafes and bars around the office, where you can pay with your badge. Yandex has its own anonymous image board, which is officially recognized. Maybe it is a plus. Most people say it is.
Cons
Salary lower than market average (it can be compensated by RSUs on higher grades, but you'll have to compete with workaholics, see below). Most of technologies are in-house: cloud, building, deploying, virtualization, monitoring, map/reduce, storage, mono-repo and so on and so on (most of technologies on market don't scale to Yandex scale). Prepare to learn from the start. If you leave -- prepare to learn again. So-called duty (dezhurstvo): prepare to support internal users, administrate your service, deal with quotas, SLBs, monitoring, fix f___ups. It can last a week (including weekends if something bad happens). And after several weeks -- repeat. There are almost no system administrators -- all developers are SRE/DevOps, whatever. Development is C++ centered (hello, building python projects into giant binary) Work/life balance is not that good -- most people around are workaholics. And you will have to compete with them, because on biannual performance review you'll have to show that your results are better. Severe Google-style interviews -- prepare for 8 hours of interviews, at least half of them are algorithms (even if you won't use any of them them in daily work). Open space environment (can vary from small offices with 8 tables to up to ~40), sometimes very noisy (you'll recognize your colleagues on the street by noise-cancelling headphones) . Many meetings, not that many conference rooms.