Treat like one of your children because that’s effectively all it is at this point.
I asked it to do something. It summarized and said do you want me to do this? I replied Yes. And then it started talking about the band yes from the 70s.
Then it kept bringing that back in later. Like I asked it for a list of something completely fucking different, and it gave me a list of yes albums. Even though I told it that yes had nothing to do with it.
I had to start a new chat. Even still wasn’t doing a fairly straight forward task properly and it used to be able to do it no problem.
2 Likes
Sound likes Rajesh & Co made one deployment too many at the weekend. The INERNET is impacted in a big way. (AWS)
No coincidence that’s it Diwali either 
4 Likes
When @flattythehurdler posted about the big job he had lined up for Friday evening I knew we were headed for trouble.
5 Likes
Failover scripts in full whack.
1 Like
Lads…I’ve a run that’s not uploading to Strava yet. This problem better be resolved soon.
1 Like
I knew I was. Didn’t finish til 630 pm.
On a Friday.
Fellas would be laughing at poor Ashutosh bating ouy updates to the code. Now imagine a robot developed by Ashutosh doing the same thing. 
3 Likes
I think we’re safe from the machines taking over for another while.
As our old stock CS lecturer pointed out, they were panicking 40yrs ago as a result of the prediction that Excel would reduce office employees to near nothing eventually.
1 Like
Better ask him to cut a spare key
TTA, TTE, TTM stats all dogshit.
1 Like
Fair play to him for fixing it in 4 minutes.
I’d imagine he wouldn’t have any Ashutosh’s reporting to him
It was a fairly limiting issue. The beauty of cloud/AWS is that you can flip over to another backup instance (possibly hosted in a different location) when you hit an issue with your live instance… however in this particular instance the Dynamo DB issue was prevent the cut over from one instance to the other.
…for all the talk and money spent on single points of failure.
1 Like
The internet is fucked again