Hi. I program advanced AI algorithms for a living. I currently work for a biotech company developing custom software. AMA about artificial intelligence or programming in general.
nice
do you have any freeware we could look at?
or what kind of things do you make?
Name:
Anonymous2013-06-22 21:25
Would you add Mentifex to your LinkedIn account?
Name:
Anonymous2013-06-22 22:16
>>2
while (! justin_bieber) {
// do nothing
}
// other stuff
>>3
All of my stuff is propriety.
I develop algorithms that determine the plausibility of DNA recombinations to survive as alternate species and other data mining operands.
>>4
The work is in confidental and cannot be shared.
Name:
Anonymous2013-06-22 22:32
What sort of applications did you write? What sorts of techniques?
>>6
I wrote many. One particular was an "aesthetic evaluator" that took large amounts of survey data according to what people thought was beautiful and attempted to determine if growth processes would result in aesthetically pleasing organisms.
The techniques we used were mostly agile programming.
Will you please make a strain of superweed. It actually doesn't have to be weed at all, something like an algae that released THC into the air or something would be nice.
Name:
Anonymous2013-06-23 1:53
Is it true that Monsanto has it's researchers assassinated if they try to quit, to keep them from giving away information?
Name:
Anonymous2013-06-23 2:08
>>14 One particular was an "aesthetic evaluator" that took large amounts of survey data according to what people thought was beautiful and attempted to determine if growth processes would result in aesthetically pleasing organisms.
To kill us neither
yet both we die
An ounce of freedom
for five whole pounds
Many words are cheap
though some cost dearly
How i lost such treasure
or how i stumbled upon it
forever after a mystery
are any AI algorithms actually good at solving problems?
Name:
Anonymous2013-06-23 10:16
About how long does it take you to program an algorithm?
Name:
Anonymous2013-06-23 11:07
>>15
Drugs are bad. You should work to remove the social pressures and problems that make you want to do drugs.
>>16
I have never worked for Monsanto. I would not surprise me.
>>18
The way I implemented a lot of algorithms used a lot of virtual functions and multiple class inheritance.
Java and Ruby are for frontends
>>25
They are good at making it look I'm doing something productive LOL.
In reality, they are useful, as when you have amounts of data of a certain size there isn't much else you can do to detect meaningful patterns above a certain level.
>>26
Depends on the complexity. A couple days for simple stuff, a couple years for complex stuff.
Name:
LeRoyand!/B832hj5qs2013-06-23 11:12
Sorry I did not put the "trip code" in earlier posts.
One project of interest I am working on is the 'trans-analytic realtime objective low level extraction daemon' which is a cluster-based background process that scours database tables up to zettabytes in size, using heuristics to determine if various threads of data are positively worth looking into. I've had to write another program to verify what it doesn't select as "interesting" is in fact not interesting. Lots of 'fuzzy logic' and stuff like that.
Name:
Anonymous2013-06-23 11:13
what's your favorite competitive algorithm
Name:
LeRoyand!/B832hj5qs2013-06-23 11:57
>>29
Mine is the "Work Function Algorithm" as proved by Marek Chrobak in the 3-server problem. I have used this in real life to add redundancy to things like what I talked about in >>28.