Pros

The work/life balance is very good, which is just a small symptom of a very relaxed culture. There is a lot of flexibility about hours, collaboration, work-style, recreation style. A lot of the engineers are of decent to high quality, so you can really learn a lot from working with people around you (this is less so than in years past as people have started to leave).

While some people are a bit cynical, no one takes the job too seriously and there is a lot of fun and good times to be had. There is a lot of room for taking the initiative and championing projects/changing methodologies/playing with new technologies. You can easily pick up a conversation in the hallway about any number of topics and people generally welcome outside opinions.

The problem space is also very intriguing (when you are allowed to work on it, see below), and there is tough competition out there that can be beat.

Cons

All that is well and good...until you get to your actual job. Orbitz moves at an extremely slow pace, and has created many hoops for you to jump through if you want to get anything accomplished. While the engineers can be upbeat, there is a lot of frustration at the fact that you can't just fix something without involving twenty people that have no knowledge about the problem or the code.

When you combine this with the fact that there is simply a lot of redundant code that has to be maintained, you create a development environment that is frustrating for engineers to work in.

The pace of releasing software is glacial, and management doesn't seem to have a clear idea why this is. Because of that, they investigate methods to try and speed it up, but without any clear definition of the problem, each solution appears to make it worse. Not only that, but each solution gets implemented only partially, which is often worse than nothing. It seems like every three months Orbitz starts looking for a new "silver bullet" to it's problems, and is unable to follow through on any change, nor able to see that it's problem is itself. Be prepared to change how you work and who you report to at any moment.

You will spend 90% of your time doing administration around your development lifecycle (coding, testing, code reviews, etc) rather than actually getting stuff done. This has caused a spiral of pain that now makes each mistake a giant blame game, and causes everyone in the company to be gunshy (not what you'd expect from a cutting edge website).

Speaking of management, over the past few years there has been a real lack of direction. That lack has allowed bad apples to enter the management forces (because under no direction, good and bad management look very similar), which is now hindering Orbitz from recovering from the various buyouts that plagued it. These people are territorial, counter productive, and often sabotage progress (either overtly or covertly). The rest are either naturally ineffective or rendered so by the political games that are played.

This has all combined to convince the best engineers to leave the company, which has left it very top-heavy, meaning each problem here becomes aggravated.

Advice to Senior Management

You need to get the ineffective and hostile leadership out of the company and give your engineers a reason to believe in the company again. Right now your best engineers are having a hard time recommending Orbitz to anyone else, even if they aren't dissatisfied enough to leave, and your worst engineers are using this opportunity to coast. You have a lot of inertia to overcome, and you need a good jumpstart to get you on your way. You've shown you can cull the herd, but you need to start culling the top.