244 karma · joined October 21, 2011
Socials: - github.com/dbousamra
---
I don't want to start a "JavaScript is bad" flamewar, but it's sad that there is an entire class of problems that we as software developers are solving, where simple type checking seems amazing
We are a home security startup, a bit like Nest and Canary. We have 20 or so developers, with a total company size ~ 50.
Tech: AWS, Scala (purely functional, none of this Akka stuff), React frontends, DynamoDB, Postgres and ElasticSearch. We have big scaling challenges (we currently service 2 billion requests a month from 10000 cameras),
We are looking specifically for Scala developers and DevOps roles. Please let me know if you want to do coffee and chat :)
More details here: - Scala: https://cammy.workable.com/jobs/363355 - DevOps: https://cammy.workable.com/jobs/365927
Disagree. Exceptions are broken in Scala. I think it's better to do what was suggested in an earlier point, and return an ADT. You should wrap them and return an ADT constructor taking in the exception.
About us: https://www.cammy.com/about/
Open positions: 2x Scala backend developer.
- We are a startup in the home security space. Cammy lets you keep an eye on your home using cheap IP cameras. Think Dropcam and Nest.
- We deal with a LOT of data (many billions of images a month) and have some fun problems to solve
- We work with a lot of Scala and FP, so if you are into that, you'll fit in.
- Have plenty of funding, and are growing almost too quickly.
Drop me a line on dom@cammy.com
That's kind of what he did with Git.
We are working on some useful features around alerting users when a camera detects movement, and you aren't in the immediate vicinity (I.e. an intruder etc), which will make it a good alternative to conventional alarms. Live view of camera feeds is also coming.
If you have a spare iPhone or Android device, you will also use those as cameras, rather than purchase IP cameras (not yet in production).
Best of all, the app is free. Any feedback?
I added an extra input to the training data for the XOR example:
net.train([{input: [0, 0], output: [0]},
{input: [0, 1], output: [1]},
{input: [1, 0], output: [1]},
{input: [1, 1], output: [0]},
{input: [3, 4], output: [7]}])
But the result of running: var output = net.run([3, 4]); // [0.987]
Is [0.9999823694382457]
which strikes me as odd. Shouldn't it be closer to 7?What do you guys think? http://www.cammy.com/