Not doing tech projects for a while...

* git is really not that simple
* CI is key
* serverless is still far superior as an approach to anything else
* most problems are people problems
* read Accelerate and listen to @nicolefv and @jezhumble

* Observability matters so listen to @mipsytipsy
* When you get to scale the stupidity of some of the decisions you make early, really come home to roost
* The most important hire is not a 10x dev but a really good VPoE
* Competence + good planning is far better than cleverness

Technology teams are often obsessed with getting great developers.

Don't.

Get great planners and architects. They're different.

You can solve most problems with great planning and system design especially if you go serverless.

But most teams still haven't figured that out.

When I've done due diligence I've often seen things that are over 5 years out of date because system flexibility is impossible so to decisions made many years ago.

Success is not a sign of good technology choice.

What makes a difference is getting your planning right, your processes right, your ability to change things rapidly and fix things fast. Those are things that affect tech choice but are often made completely separate to programming language or framework or infrastructure choice.

I feel like the tech world is often years away from learning this broadly. I know lots of people know this but it's not many in the grand scheme of things.

And many will say "oh you can do this with <insert tech here>" and that completely misses the point.

If your hill is "my tech is best" I basically stop listening.

Serverless is an approach not a technology.

What I'm interested in is your approach not your tech.

Almost every problem is a people problem.

Serverless your people.