Unfortunately, the talk didn't work for me. While the presentation style was fine, I was hoping to pick up some tips and techniques for handling actual legacy code, especially the old code where concerns aren't properly seperated.
The talk seemed to be mostly about handling corporate culture around large-scale projects with a fairly well organized code base. I'm sure this is very useful for people in that situation, but as a developer of small-scale client-driven projects there was little in it for me.
Very insightful talk. I had zero exposure to cloud-based hosting coming in to the talk, and found it easy to understand the various concepts and considerations explained. You showed a lot of practices that seem pretty obvious choices when you explain them, but which would very likely have tripped me up when starting a cloud project without this talk.
Pleasant presentation with a nice mix of showcasing performance, tips for getting your code ready, and tips for getting the most out of your server environment. Very useful to hear before starting our own deployment. Only minus would be that the talk lingered on some minor implementation details a bit long sometimes.
Great presenting style, and very informative. Most of the advice seems to be very real-world practical, and there's a lot you can pick up from this talk even if your application doesn't require the full rigor of a public API.
I joined this talk expecting it to be somewhat out of my comfort zone, not having touched any internals and not doing any C for the past 15 years. While I was indeed out of my depth, I really enjoyed the in-depth technical level, and I was happy to find that I could still follow everything quite clearly. You're clearly very comfortable with the subject matter, and present it in a very natural and non-intimidating fashion. More hardcore-tech talks like this please!
The presentation was done well, but the talk lacked some depth to me. Mariusz Gil's talk on friday went a bit more clearly into the mechanics of the data prep and use of the learning machines.
Well-presented and with just enough depth to show where machine learning could be useful, and where the tricky parts are (data prep).
It took me a while to understand what the real topic was but it was a nice presentation.
Nice presentation. Good speaker and interesting topic! Too bad there was little time left at the end. It looked to me like there were more people with questions and there was a beginning of what could have been an interesting discussion about commands.
I liked your talk and you presented it enthusiastically, however I did get lost a couple of times and it took me a while to get the message (might have been due to tiredness).
The part where you introduced some scala concepts was a bit too rushed making it hard to see the advantages.
The ideas of your talk were definitely interesting and you presented it very well, but I will need to do some additional reading before understanding the bigger picture.
(On that note do you have some references that discuss the advantages and disadvantages of eager design?)