Never Trust a Programmer

In the agency world (where I was), carefully billing time was de rigueur for client-facing people. I pushed to have it apply for non-billable internal people too.

So the "idea people" had to do cost-benefit. Requires a cultural change (and trust), but it brings discipline to both sides. Really helped.

Indeed; my worse experiences have been exactly of this sort.

People say they want to be agile, but they rarely realize that it also means you have to do less.

I often got hammered on not being fast enough, and yet, when cutting corners and throwing it together with boogers and duct tape to get it to a demonstrable state, I was told it wasn't presentable to customers because it didn't have the frills and throw pillows.

Next time, I'm just not working with someone that doesn't understand mvp.

The sales and PMs would promise the world to the clients, and development hours were taken away from the developers to support the IAs that were overbudget. We had a solid dev team and delivered the sites on time, but one of the projects was all wrong in the client's eyes (due to poor specification, the whole IA -> Design -> HTML workflow was a very slow, behind schedule waterfall) and it was blamed on development. We quickly made the changes, that project ended well.

I ended up leaving after my co-op to accepted an offer from a prior co-op. Shortly after, most of the dev team quit, and the VP of technology was fired. A couple of months ago the company closed its doors, owing months of paychecks to employees, and clients lost project/money.

Avoid companies like the one described in this article like the plague.

(Although I suppose it's hard to tell what things are like from the interview process alone, but a business model based on selling to large enterprises is a red flag.)

Necessarily a manager evaluates you by whether the product works or it doesn't and is on time or late. You are evaluated against management expectations. The more that they expect,the worse you do. Because expectations are typically unrealistic,you can get a fine reputation for perfunctory but consistently on time work.

Decrease early expectations to do well. During project planning be pessimistic and extreme humble. Programing and research have very high product and deadline risk. For each project have a contract with your manager that identifies the main risks, and accept responsibility for only what you can do something about. Your professional self management will impress your manager, but he will hate the uncertainty.

Happens way too often. But it's understandably difficult to solve as an organizational issue.