News for hackers. Stuff that matters.

zintinio5

Re:Tech takes over: New York is the sector's second city on Monday February 26, 2018 @03:15PM (#16468582)
The job description is to-the-point and doesn't ask for anything out of the ordinary. Other than mentioning a CS degree, I give it two-thumbs up.

Hiring is definitely a tough nut to crack. Personal referrals are by far the best source to get solid people quickly, but it's also a finite pool. Recruiters are definitely a mixed bag. I would suggest also sending a representative to a job fair (face-to-face still works!). It sounds like you're doing the best you can besides creating some really unique opportunity.

We had some luck bringing in people interested in functional programming, but ended up having to go heavy on junior devs since that's what's easy to hire for. We had some mid-level and a couple senior people which were hard to come by.

It's going to be a grind no matter how you approach it. All I can suggest is to make a really compelling offer, technologically, financially, or otherwise. Also consider finding older developers interested in part-time or remote work.

Re:Tech takes over: New York is the sector's second city on Monday February 26, 2018 @12:34PM (#16467112)
First, what does your careers page look like? Sometimes the job description is unclear or you might be filtering out qualified people. Second, mid-senior is really tough to hire for, I just experienced it first-hand while hiring for my last employer. Regardless of experience level, you are forced to compete on two fronts: interesting work, or high compensation. Mid-senior devs can afford to be very picky and if something seems a little bit off, they'll look elsewhere.

Have you partnered yet with a recruiting company? I personally just went through TripleByte and they could use a few more NYC partners. They do their own pre-selection of candidates to save everyone involved a lot of time. Another good one is FunctionalWorks. Meetups are also pretty good for making organic connections with higher-quality devs. Every time I go to a tech meetup at least 3 or 4 employers are represented and hiring.

Re:Tech takes over: New York is the sector's second city on Monday February 26, 2018 @12:00PM (#16466802)
That's not the case everywhere. Some companies pay managers less than engineers. From the company's perspective, a manager deals with a larger chunk of the business, while an engineer solves just one small part. Not saying it's right, but that's my simple guess.
Re:Tech takes over: New York is the sector's second city on Monday February 26, 2018 @11:54AM (#16466745)
We're pretty comfy here 9/12 months a year. The Spring and Fall are fantastic, and Winter is getting milder each year.
Re:What to do when tech jobs go bad on Monday February 19, 2018 @11:44AM (#16413896)
I would quit this team.
Re:Ask HN: What has HN given you? on Monday February 19, 2018 @01:42AM (#16411156)
Coming from an under-represented demographic in technology, Hackernews has really been my insight into the industry and helped me immensely with figuring out how to get into it. Reading this website has honestly made me a ton of money and steered my career. On the job, I think Hackernews also keeps me sane by stimulating continuous learning each and every day. I really like this little community.
Re:Firefox 58 on Tuesday January 23, 2018 @05:32PM (#16218484)
Thanks! I was browsing a few days ago and applied to work on Telemetry in Firefox and ML Engineering at Pocket. A lot of the work being done with respect to the Quantam initiatives seems interesting, but I didn't really see anything related to the core browser. Anyway, thanks a lot for the info.
Re:Firefox 58 on Tuesday January 23, 2018 @02:38PM (#16216817)
How does one end up in a position to do this work at Mozilla? I'm currently looking for a new SWE role, would absolutely love to work on Firefox.
Re:Ask HN: Who hires mathematicians? on Thursday December 07, 2017 @08:48PM (#15876085)
Still the same situation, even for CS majors from less known schools. Somebody has to take a chance, few people want to take a risk on you.
Re:Ask HN: Who hires mathematicians? on Thursday December 07, 2017 @08:45PM (#15876073)
I will keep this in mind when researching jobs.
Re:Ask HN: Who hires mathematicians? on Thursday December 07, 2017 @06:46AM (#15869265)
I've seen at least one or two companies working on optimization at least here on HN. I'd take a guess that Gurobi is one option, and I saw SigOpt post a listing on the recent "Who's Hiring?" thread. Google seems to also do a lot of optimization related work as well. I'm in the same boat, though I was stronger in CS during my studies. The background is useful, but you will only ever use bits and pieces of it at opportune moments.
Re:Ask HN: Who hires mathematicians? on Thursday December 07, 2017 @06:42AM (#15869244)
Any chance you know someone hiring for said scientific computing roles? I studied Applied Math / CS undergrad, have been gearing up to work in Autonomous Vehicles, but really any challenging numerical work would be interesting to me.
Re:A Guide to Natural Language Processing on Wednesday November 15, 2017 @03:29PM (#15707447)
Like everything else, depends on your use-case. I have personally used TF-IDF vectors and token sets with Cosine and Jaccard distances in practice.

Some examples of use-cases: are you searching for "semantically similar", or "near duplicate"? You can compare documents under different metrics and different _representations_. Some representations are: LSA, PLSA, LDA, TF-IDF, and Set representations, along with metrics such as Jaccard Distance, Cosine Distance, Euclidean distance, etc.

Doc2vec is the Word2vec analog for documents.

Re:The often-overlooked random forest kernel on Sunday October 29, 2017 @11:28AM (#15579632)
This is really awesome, thanks for posting! I was not yet aware that random forests could be used as kernels, and the link to Davies and Ghahramani is especially useful, definitely will read through later.

Ask HN: How to avoid being typecast as an engineer?

Hi HN, I'm currently in an employment dilemma where I want to move into an ML engineer or Data Scientist role at another company, but I've spent the last 3 years as a Data Engineer and other companies seem only to want to hire me for that. Previously I've worked on web and mobile development.I've tried joining a very early startup to gain the social capital to work on data analysis and machine...

( Read More... | no comments | on HN )
The problem is distributed training, not inference.
There might not be a need to do so at the moment. Otherwise, they might open that up at a later time.
Re:End to End Machine Learning Pipeline Tutorial on Monday July 17, 2017 @11:30AM (#14789456)
I'm gonna go through it as well, that was a significant amount of work you put in.
Re:Monte Carlo theory, methods and examples (2013) on Friday July 07, 2017 @09:40AM (#14718213)
I certainly have in the past when something really grabbed my attention. My library has expanded in recent years partly because of the topics shared on Hackernews and mentioned in discussions. I would hazard a guess that most people don't follow up and just upvote a cool sounding title. It really does take months/years to work through some of this material, but there are some people who do it. Also important to keep in mind is that readers specialize (I'm big into math/stats/ml, but some of the guys I work with are deep into logic programming and theorem provers) so no single reader is knowledgeable in everything posted here.
Re:Fantastic DSLs and where to find them on Thursday June 29, 2017 @11:39AM (#14664135)
Immutability is absolutely fantastic as a default (especially for typical business logic). I'm not sure I want to go back to mutability except in constrained environments or time-sensitive software. Mutability should be harder to reach for because it forces you to think hard about the implications to other parts of your program. I can think of a number of times we've had to parallelize a routine or share resources between workers, and it has been almost trivial _because_ of immutability. In contrast to that, I inherited a hairball of C code using pthreads, mutexes, global variables, etc which had to be completely gutted and rewritten as it was impossible to understand the dependencies in the software (we were witnessing segfaults, deadlock, etc).
Re:Show HN: An improved version of HyperLogLog on Wednesday June 28, 2017 @03:31PM (#14657455)
Neustar Research has a good article on this which may be of interest to you. However, I think the approach they took was still based on the inclusion-exclusion principle (still a good read though).

* https://research.neustar.biz/2012/12/17/hll-intersections-2/

Another approach if you have the flexibility of changing your implementation, is to use a different type of data-sketch that is more amenable to set expressions. There is some discussion (and references) here:

* https://datasketches.github.io/docs/Theta/ThetaSketchFramewo...

* https://datasketches.github.io/docs/Tuple/TupleOverview.html

Re:Show HN: An improved version of HyperLogLog on Monday June 26, 2017 @01:25PM (#14638597)
If what you want is an approximate answer, take a look at

* Counting Bloom Filter https://en.wikipedia.org/wiki/Bloom_filter#Counting_filters

* Count-Min Sketch https://en.wikipedia.org/wiki/Count%E2%80%93min_sketch

Here's a library: https://datasketches.github.io/

However, you can probably fit an exact answer into memory with just a simple hashtable if you've only got 20 million objects. A single byte could be used as the state marker.

Either that, or just track 6 counts.

Re:Clojure Newbie Guide (2015) on Thursday June 22, 2017 @05:08PM (#14615382)
There can be problems with state management, which I've only seen with our web application. Sometimes a hard stop / start is what you need unless you designed the application for reload-ability. We're working towards that for our web application. Frameworks especially seem to exacerbate this problem, since you have limited control over the execution of your program.
Re:Clojure Newbie Guide (2015) on Thursday June 22, 2017 @09:39AM (#14612124)
I've wrapped C and C++ code via the JNI and called it from Clojure. There are a good number of projects wrapping C code in Clojure. It's a little awkward (Clojure -> Java -> JNI -> C), but works well enough.

We get around the startup time by keeping our work either in the REPL, or just having large enough tasks that the loading is dwarfed by the time it takes to get started. Webdev might be a little crappy if you're used to the turn-around cycle of either node or PHP.

I worked on this as part of a project with US Fish and Wildlife in 2015 for bats! We were monitoring local populations to track the spread of White Nose syndrome. It wasn't in 3D, but we linked the audio data / classifications with lat/long points and plotted population estimates on maps.
Re:Regex: badly needs fuzzing on Wednesday February 08, 2017 @04:36PM (#13602370)
Ah, fair enough.
Re:Regex: badly needs fuzzing on Wednesday February 08, 2017 @03:35PM (#13601794)
From my understanding the main benefit of RE2 is not speed, but linearly scaling execution time with respect to the input size, along with bounded memory usage. For certain inputs, it may outperform other engines, but the converse may also be true. As with any feature that may be abused, backtracking may also be useful: for example, you may need to write a script to munge text. Since you're not exposing it to arbitrary user input, it's a reasonable feature.
Re:Announcing GVFS: Git Virtual File System on Friday February 03, 2017 @08:08PM (#13564925)
Several years ago a friend and I had a need to build a virtual file system that was portable between Linux and Windows. At the very least, we attempted to share as much code as possible. It proved to be pretty easy and we had a working prototype after about 30 hours of work. We used FUSE on Linux, and Dokan FUSE on Windows.

[1] https://dokan-dev.github.io/

[2] https://github.com/dokan-dev/dokany/wiki/FUSE

Re:Ask HN: Who is hiring? (February 2017) on Wednesday February 01, 2017 @02:53PM (#13544363)
It's consistent with what people are posting and the directions given. From the very top:

> Please lead with the location of the position and include the keywords REMOTE, INTERNS and/or VISA when the corresponding sort of candidate is welcome. When remote work is not an option, please include ONSITE.

Re:Ask HN: Who is hiring? (February 2017) on Wednesday February 01, 2017 @12:25PM (#13542703)
You need to account for ONSITE. That's specified at the top, but I don't see it anywhere in your spec.

Older »