LinkedIn

1. Engineers are pretty good.
2. People care about the site and what they are working on.
3. Performance is (usually) rewarded.
4. You have a huge impact on the world.
5. The day-to-day environment is pretty nice.
6. There are interesting problems to solve in architecture, scalability, etc.

1. The architecture of the site is in shambles. There are over 300 distinct services on the site. Not a single person could draw even a quarter of the current architecture.

2. The codebase is in shambles. There are no comments or documentation in any of the codebase. The wiki may be years out of date.

3. Within the last year, the culture here has gotten extremely political. I'm afraid to do a great job and discuss what I'm working on because it may cause a turf war.

4. Speaking of culture, LinkedIn's is a "culture of fear." We are over post-mortemed. People have responded by going through ridiculous lengths to CYA.

5. Middle management is highly variable in quality. Middle managers are promoted by how well they manage upwards, not how well their teams are doing. I've seen people promoted that I frankly felt should have been shown the door based on their lack of professionalism and performance.

6. Engineering is stuck in a vice grip from product.

7. When a team is mandated to build a piece of infrastructure, it takes forever and usually comes out as an over-engineered mess that's less capable than the open-source alternative they spurned to write it in the first place.

We need about 10 more inversions.