Hacker News new | comments | show | ask | jobs | submit login
Ask HN: What did you work on in 2017?
350 points by freeelncer 10 months ago | hide | past | web | favorite | 562 comments



I quit my job as a software engineer at Google early this year to teach people how to code. I started paying people $15/hr to learn so they can make ends meet while learning instead of working at Walmart.

I thought about all the missing pieces in my engineering growth and created a curriculum that welcomes students from 0 engineering background and plugs in all the holes that were black boxed to me in my engineering growth: We host our own servers, allowing students configure nginx and create ssl certs themselves for the apps they build. Our projects mimick existing well known companies (netflix, dropbox, gmail, google docs clones).

Our curriculum is largely project based, so students work together on projects that they would be using themselves: building their own email client, chat client, filestorage/backups, firebase, etc. From day 1 of a students journey, their code is thoroughly code reviewed by other students.

2 months ago, Calworks, a local government assistance program, offered to send students to us and pay each students $13/hr for up to 6 months. Unfortunately, to make this deal work, we needed a commercial office (my wife and I teach out of our apartment) and we did not have the financial resources.

Last month, we finally got approved as a tax exempt non-profit so I can reach out to my friends for donations (but donations take time, I have to set up a bunch of fundraising tools first). My savings ran out so I started applying for jobs and landed a full-time position at Paypal starting in January.

Moving forward into 2018, a few of the senior students are going to be leading the non profit. 100% of my salary and equity is going into the non-profit so existing students would not only continue to be paid, but we now also have the financial resources to get an office and push the Calworks deal through to help more people! 2018 is looking to be a great year.

We do not have any internet presence at the moment because this year our focus had largely been testing and iterating our curriculum as well as our financial model. 2018 will be different and if you want to help, our non-profit is called GarageScript.

https://www.facebook.com/garagescript/


While your post is getting lots of reads, you should set up a landing page to collect email addresses. That way, when you have your non-profit set up, you send an email out asking for donations.


This is a great idea! Adding this now.


This is absolutely awesome. Sounds like a really great way to make the world a better place.

Not everybody is using fb, I hope you'll finish your http://garagescript.org/ website soon because now most links don't seem to be working. When you provide more details I think you may find some people who are willing to donate. E.g. is non-profit registered under garagescript? I couldn't find it. Well at least make contact and linkedin links working now that you are going to get some exposure from HN frontpage.

Got a bit too much into details, but what I meant to say, really great initiative, kudos.


Thank you! Non-profit is registered under 'garage script', here is the info on GuideStar: https://www.guidestar.org/profile/82-3129102

All of the points you mentioned is in our top priority list for early 2018, thank you for pointing them out!


> all the holes that were black boxed to me in my engineering growth

What were the holes that black-boxed you in your engineering growth?


Here are a few off the top of my head.

1. Servers. I've always 'pushed to heroku' without really knowing how servers worked.

2. Git. I've always just used gitlab or other git hosting services without knowing how to build it.

3. File hosting. I've always just uploaded files without knowing how files are processed / stored.

4. Email. SMTP is like a big unknown, I've never really cared about how emails worked.

5. SSL. This was always done for me, I've never had to create and manage my own certificates.


i admire what you're doing but I wonder why spend any significant amount of time on most if not all points on this list? if the goal is "couch -> working developer" then I don't think they're relevant to getting ramped up and working as a programmer/dec/swe who can deliver value. the black boxes you talk about might be better off taught in a lazy fashion so to speak.

a lot of these seem devopsy. I've definitely delved into a lot of these as an extracurricular activity over the years or have had to learn for work reasons (like being the only guy willing to take on the devops type work) but I'm not sure it belongs in a boot camp type curriculum.


You are right, we actually don't put much effort teaching these things, its pretty much lazy learning when it comes to devops. However, I make it a point to host and handle everything in-house so students see the whole picture of how the internet works. The alternative is to push to heroku or host code on github, which introduces black box.


What you're doing sounds great. I'm on the board of HackerDojo, our mission sounds pretty aligned to yours and you are not too far away - so get in touch, I'm sure we can figure out a way to support what you're doing! (will send pm also)

https://hackerdojo.com/


I would love to donate time to build you a website/online presence if you need it. email is in my profile if interested!


Responded, I think donating time is the best form of donation. Students want mentorship from industry veterans but its hard to find people willing to commit the time to mentor. Thank you for offering!


Making their own pools reminds me of PARC, and minimizing black boxes/assumptions must be very satisfying.

There's so many assumptions in mathematics, I've always assumed it must just take too much time/expertise to cover them properly. Engineering != math, but accomplishing that is revolutionary.


You know Wo/Man I think you should put up a page and ask for people to donate as sponsors. I think HN users alone, if we come together, could send 10,20,50,100 people through your program. I would be proud to know my few dollars helped in this way.


Yeah! Thats the plan for next year. It takes time to set up a donation platform as an NPO and we are working on it. In the meantime, I put an email signup (per baron816's suggestion) and will send out an email when donation tools are ready).


I would like to put some of my time helping your initiative. Do let me know. Email in my bio


Thank you, reached out.


Hey, I tried doing something similar in my city. I found designing a useful curriculum a challenge. I would love to ask a few questions. Couldn't get hold of your email. My email is raghav.toshniwal at google's service.

Thanks!


Love to give you some feedback/pointers, shot you an email. We plan to open source our curriculum and make adjustments as we go :)


Would love to donate to such projects (and already have, but you sound neat too).


Thanks for the support! Would love to share donation link but it has not been set up yet :)


If at all possible, let me respectfully suggest you make that a priority. I realize with an NPO it is a bit more complicated than for an individual.


Thank you, I'm on the same page. And you are correct that NPO takes longer and is a more complicated process, but we are diligently working to make it happen.


(Moral of story: don't announce genuinely amazing new initiatives without taking the time to get the donation links working first!)

Curious how/why something like PayPal/Stripe integration couldn't be done in a short amount of time. Not as a criticism, just "is this actually technically hard?" (or is it just a time issue?)


1. I was genuinely sharing what I did this year, was not hoping for any positive financial outcomes.

2. Integration with non-profits donations is not trivial. After registering as a non-profit, it takes some time for the list of registered non-profits to propagate through to donation tools (facebook, gofundme, etc). Sometimes, I have to fax in documents to prove legitimacy, and it takes a few days for that to process. I could easily create a gofundme for "help me do x" in my name instead of the non-profit's, but I didn't want to do that.

Hope that clears things up.


Check out givelively.org for (free) Stripe-based nonprofit fundraising and payment tools. I used to work there and they are doing a great job simplifying the process. I'm happy to put you in touch with the team if it's helpful.


givelively.org is amazing. Signed up and would love to be put in touch. song garagescript org


I see. Thanks for the clarification, very interesting.


This is awesome! Please keep us posted on where we can donate in the near future.


Will do! I'll respond back to this thread when things are all set up.


Its just like Udacity.


I quit what many would consider a successfull job in programming and had a sabbatical year. Moved to a smaller town and down sized everything in spending to the point where I pay $600 per month for housing, food and bills, so living of my savings haven't been an issue at all.

I have focused on things like reading (read +40 books in 2017, up from 1-2 per year), wood working, sketching, running and skiing. To keep up my programming skills I have done a deep dive in new programming languages and fiddling with some side projects. It has been an incredible year for personal development and it has changed my perspective on what things are important in life (sitting in front of a screen 40-60 hours a week not being one of them). I highly recommend everyone to do this at least once in your career!


What did you do about your social life? Isn’t that lonely to be away from friends and family for so long?


actually I moved back to friends and family. But it does get lonely sometimes when you don't have a strict schedule and lots of free time when people are locked in at their 9-5 jobs..


I'm 27 and I did the similar thing last year. Having plenty of time for personal development is indeed great. But to be honest, I'm anxious about the limited social networking in smaller town from time to time.


> I pay $600 per month for housing, food and bills

Is that in the US? Sounds really low...


>smaller town and down sized everything

You can very much pull that off in lots of small towns in the US, but yeah, you're living a close to thread bare lifestyle.


Not really. My girlfriend and I together make more than 200k/y and spend less than 1k/month on expenses while we prepare for her to launch a business. We cook food from CSA and our Farmer's market, have a cozy 2 bedroom in a quiet city and generally just don't buy a lot of things we don't need.


Which city? And 1k includes rent?


it's incredibly how much you can spend on take out, drinks and useless knicknack/gadgets. It was easily $300-800 for me before (camera geek).

But made an active choice to cut them out. And it hasn't been a loss for me, instead of take out I make exactly what I want (and no stress now that I have time to cook), instead of drinks out invite friends over and have a drink (cheap) at home, instead of useless things try to realize that you don't actually need them. Also living in a place with public transport helps, but I mostly bike everywhere.


Nope, in Sweden


Hi I've also recently (re)discovered reading books. What books do you recommend?


I use GoodReads a lot for discovery. See what friends are reading and browse lists that people make (https://www.goodreads.com/list)


Curious if you'd like to share your age and when you started your professional career? Do you recommend doing this earlier or later in one's career?


I am in my mid 30s, started my professional career in 2008. Hard to tell if when is a good time for doing it. My feeling is I should have done it earlier but I guess it depends on your situation. Maybe if you have an active social life and hobbies you love it will be easier at any point. If you have worked a lot and over long time I think friends/hobbies can fade, and it will be easier to fall into loneliness/depression if you quit working.

But, I think doing it early (say after ~3 years) has the advantage of settling/deepening the understanding of your work. Often the early years are super hectic and you try to be your best, neglecting things like mindful reflection. I believe sabbatical can help this, kind of like how sleep organizes and "cleans up" thoughts/memories. We need both the short sleep and "long" sleep.


Do you mind sharing the town?


Trying to be the world's first person to be cured from autism. I believe my autism is caused by the shape of my forehead, which puts too much pressure on my muscles. Evidence is here: https://corticalchauvinism.com/2017/11/13/yuval-levental-cra...

Also, I have prepared for a potential surgery by getting botox injections in my forehead muscle. So far, my focus at work has dramatically improved: https://corticalchauvinism.com/2016/10/17/yuval-levental-aut...


Yuval, are you familiar with Barry Neil Kauffman’s work? Specifically, his book Son Rise. It tells the story of how he and his wife cured their son of his autism.

https://www.amazon.com/Son-Rise-Barry-Neil-Kaufman/dp/091581...


I have heard about it, but I'm not sure it would help me. Thank you anyway though.


From all of the research I've read, autism seems to be a neurological disorder. Is this not the case? Genuinely curious what led you down this road rather than others as I've been struggling to reason about similar issues and tendencies I've had.


What I am trying to do is change the contour of my forehead muscles so they don't feel as intense. The forehead muscle has nerves, so it is part of the nervous system.

What led me down this road is I found a study that said that people on the spectrum may have different facial features (https://www.cbsnews.com/news/children-with-autism-have-disti...). I reasoned my face looks very similar, and then realized that I experienced lots of muscle tension on my forehead.


In my work I identified roughly 16 major drivers of ASD, but most of them tend to manifest as a GABA/Glutmate imbalance or GABA dysfunction.

Lots of people have been cured depending on if their ASD was caused by biological dysfunction - for example I know one who had biontidinase dysfunction so biotin supplementation cured all symptoms.

Hopefully you’ve identified the driver of your ASD and it will help you achieve your goals!

I learned to deal with my personal Aspergers but despite the social issues it caused earlier in my life deeply value the weird thinking processes it has given me.


Interesting. I know an autistic person with cerebral folate deficiency who takes medication for it, and it has significantly improved his symptoms. I wouldn't call it a permanent cure, however.

As mentioned, I believe the driver is my facial structure. I know that normally, facial structure is not the cause of a condition, but the muscle tension on my forehead is caused by my facial structure, and I suspect that muscle tension is most likely the cause of the symptoms I face.


This is fascinating!

The link in your article no longer works (http://www.abaphysicaltherapy.com/2011/10/what-is-craniosacr...).

I really like your analytical approach to solving this, e.g. https://corticalchauvinism.com/2017/11/13/yuval-levental-cra...

Most people suffering from this condition don't have a physics background and wouldn't think to apply Hooke's law here.


Thank you :). The external link was only to source one of the images, so I'm not sure if I should bother replacing it.

I actually know quite a lot about math and science, but my major difficulty is not being able to hold down a career where I have to apply math/science (though I have difficulty staying focused at most jobs that require substantial abstract ability).

In addition to Hooke's law, I simply felt a lot of muscle tension on my forehead growing up. Because I essentially had that feeling from birth though, I didn't consciously notice it until recently.


Interesting. Have you found a difference in your sleep patterns since getting the treatment? One anecdote I've heard is that people on the spectrum have difficulty entering REM sleep[1]. Since REM sleep is partially initiated by eye movement, it makes sense that forehead and sinus shape could impact it.

[1]https://www.amazon.com/Why-We-Sleep-Unlocking-Dreams/dp/1501...


This sounds cool! You seem high functioning. I know someone who's low functioning, about grade 2-3 mentality and can't keep conversations going for long. No savant level skills either. How does your work affect people on that end of the spectrum? Do you see any of your work applying to them? Sorry if you've answered any of this in your research. I'll check your site out more tomorrow.


It will have little to no impact on low-functioning autism, unfortunately :(. Researching low functioning autism is far more complicated than the type of research I am doing.

BTW, I am diagnosed as high-functioning, but I am not so high-functioning that I can "pass" for normal or even close to normal in public. That is starting to change though.


Love to learn more about this. Super neat.


Left $150k software engineering job (well got fired for making video called "9 Ways to cope with having a boring 9-5 job" which somebody found and sent to HR) to make videos about stuff I think is interesting (https://www.youtube.com/c/JDiculous1 https://www.facebook.com/HonestLogic), with a slant towards addressing wage slavery, basic income, student loans, capitalism, etc. Still in the early stages, but I'll be hitting this hard in 2018.


You will get the experiences you need to go where you're going; being fired for pointing out obvious is not that bad, you did nothing wrong, it's their shitty karma. Can't you see the irony in doing something like that on facebook? The solution starts with making the right choices yourself.


Sounds like wonderful co-workers to have. Probably a matter of time before being stabbed in the back in such a work environment.


Annoying that somebody rolled on you for that.


Yea it's pretty lame seeing as how the video was posted under a pseudonym, meaning that person did some serious digging and clearly had an agenda (I'm 95% sure who it was). It's all good though, I had been wanting to leave for a long time but had kept putting it off for the "right moment". I'm glad that that decision was made for me because now there's nothing to regret.


You must have pissed that person off at some point. Without knowing the context though, that's a very low road move to take on their part.

In any event, I enjoyed the couple of videos of yours I watched.


Your Twitter is under your real name and in the About section though. Was it there when you got fired?


Wow, is it even legal in your country? What was the official reason mentioned? Maybe you can sue them?


In the US this would be 100% legal. For almost all jobs, you can fire someone for any reason (I don't like your shirt, your jokes weren't funny, it's a Tuesday) or literally no reason at all. The exceptions are few and far between - unions provide additional protection via collective bargaining agreements; you can't fire someone for being in a protected class (i.e. because of their race or gender); and you can't fire someone in retaliation for something like a wage complaint. You could almost certainly fire someone for making a video you don't like (unless they're in a union, which is very few tech employees.)


While it is true that most employment is “at-will” (which means that both sides voluntarily choose to employ/be employed) for both sides, in a friendly employee state like California, there are numerous wage claims and employment law suits that come with doing business. It’s not as easy to fire in some states.


Yea it's legal here in the U.S. since it was at-will employment. They tried to play down the effect of the video and made up some vague performance-related reasons.

It's all good though, I was planning to leave anyways.


Can you link to the vid that got you fired?



Videos are pretty funny! Keep it up :D


Did you speak to a lawyer? Firing you for that sounds illegal.


It was at-all employment so they can fire me for any reason here in the USA.


That’s not quite true. There are still illegal reasons for firing (discrimination, retaliation for whistleblowing, etc) though fighting back can be difficult and costly. But if you’re okay with the current state of affairs (I was with my own previous incident - the company was awful) it can make sense to just move on.


That's true. I should've mentioned that they tried to play down the effect of the video and made up some vague performance reasons. But yea I'm fine with it and had been planning to leave for a while, so no hard feelings. The best retaliation would be me achieving success in my new endeavors.


Awesome videos man! Resonated with me since I live in the East Village as well


That's ridiculous.


Quitting my full-time job to pursue my side-projects was the best thing I could have done for my health and sanity this year.

I am now working on a bunch of ideas that I hope will help some people around here:

1. A Pocket-to-Kindle service that syncs (almost) instantly to your Kindle whatever article you save, formats it like a professionally edited book, cleans up ads and takes advantage of the new typesetting engine inside the new Kindle firmware.

2. A Spotify music discovery website. I'm trying to make a two-click-playlist-generator by using Spotify APIs to look at the top artists/genres of a user and create playlists on the fly with tracks that the user could like.

I use Spotify daily and found myself overwhelmed by how much music there is available. Because of that, I'm mostly listening to my saved songs, Discover Weekly/Release Radar and trying out playlists that usually have the same too popular songs.

3. An adaptive brightness/contrast app for external monitors. Adjusting brightness using the monitor's controls is always annoying to do.

4. A morning alarm that starts playing an algorithmically generated Spotify playlist each time, with fade-up volume, external speaker support, adaptive algorithm based on likes/dislikes and self-updating alarm times based on day moments (twilight, sunrise, golden hour, dusk etc.)

5. A detector for processes that eat up all your CPU and battery. I started writing this in Rust so I can make it cross-platform and learn the language at a lower level.


I just recently joined Spotify. I have found their "we will play similar music after your music ends" feature to enable me to discover lots of new artists. I find it interesting you found Spotify lacking here, because I am finding the opposite.


I'm a long time Spotify user. I love that feature too and it worked very well for me when it first launched. But after a while it started playing the same songs that I have already heard many times. Spotify's algorithm is very unpredictable so I can't say it will work out the same for you. But if it will, at least you can have another try with what I'm trying to build ^_^


Gotcha. That makes sense. I could see how over time that feature gets less useful. Looking forward to trying your tool, if you have a website or mailing list for it, please share.


I agree! I love this feature. It's so good that I sometimes don't even notice that my music ended.


I found it ate so much data that I had to turn it off :/


Dude, LINKS! My mate was looking for almost that exact spotify thing!


None of those ideas are ready for public use yet. I mean they work, I use them daily, but the interface for non-programmers is still in the works.

https://github.com/alin23/spfy is the core of that idea. If you want I can help you or your friend set it up. Or I can let you know when the finished website is ready.


Ad 1. I am really interested of about the progress of development of your service. I would love to see it as "Show HN" submission.

Personally, I am thinking about developing an online service, which will be offerring general-purpose providing distribution of content to Kindle or other ebook-readers. Initially, I thought about creating only a app which will extract a article tag from website and create ebook from it (pandoc) or exactly a Pocket-to-Kindle. But when I was thinking more about it more ideas had come to my mind, HN-frontpage scrapper, simple notebook/orgmode adapted to Kindle Experimental Browser, markdown-files-to-ebook, arxiv-to-ebook converter, some kind of IFFTT pipeline (stream X from Y and save it as a ebook) and even trying to implement most of features known from Calibre application... If I could somehow help you as a developer or you are interested in bootstrapping some online service together I really would like to receive an e-mail from you. My e-mail address: (put my HN nickname here) at gmail.com


Was looking for same Poket->Kindle app, found nothing appropriate. Spent some time googling to gave up with idea of building such app. I thought nobody use kindle to read web this days. Solve my problem with push-to-kindle browser extension and manual file copying.


I would consider offering Instapaper as well as Pocket. I generally find Instapaper does a better job overall.

The number of times Pocket just either redirects to the original article, or drops salient bits of text (in particular, unordered list items) is really annoying.


It already works with Instapaper too, I just explain it as a Pocket to Kindle service because that's what people use the most these days.

Personally I use Instapaper more for reading because of all the premium features being free and keep Pocket in sync using IFTTT for their good recommendations.


What did you do to prepare for leaving your job? Do you have another source of income?


I don't have any other source of income. I had a pretty good paying job for the past year and managed to save one year worth of savings. I can survive with what I have until June (I think). I'm doing my best to create something that helps at least a few people, and maybe get some money from that too. Even if I don't, I have already learned and practiced so much more skills than I could if I had a full time job and that makes it worth it.


I would pay for that pocket-to-kindle service


I would let you pay for it if it was ready

I can make you an (unofficial, free) beta tester if you want. I'll have most of the basic functionality by the end of January. If you're interested let me know here: alin.p32@gmail.com


Instapaper have a send to Kindle option on a daily or weekly frequency.


Have you tried https://p2k.co/?


Yes, I've used it for a few months. That's the reason I've started working on my thing. I felt that it didn't have the features that I needed the most (hyphenation, dropcaps, ad removal etc.) and I would have felt like a jerk to ask for these things from the developer. These are not common needs ¯\_(ツ)_/¯


Nice and good luck!

I've started working on a similar service but didn't ship. I'd love to chat (email in profile).


They're my needs haha, I sent you an email to become a beta tester


do you need any more beta testers?


Hosted Comments https://www.hostedcomments.com/ , a Disqus alternative with a focus on privacy. The learning experience of building Hosted Comments was great : using iframes to embed comments in websites, building a commenting system with voting and some features which Disqus does not have : locking comments, hiding comments (not yet deployed https://imgur.com/a/R89Cw ). It started out as a sideproject, then decided to go the SaaS route and now a little confused about whether I want to pursue this as my main project. I'm thinking about releasing an open source self hosted version and continue offering a managed service.

Bored Hackers https://www.boredhackers.com : a public chatroom based community site. Think of it like reddit, but chatrooms insteads of forums. I just deployed the first version a few hours ago. Bored Hackers is an experiment at building the community site that I wish existed : public chatroom based communites, pseudonymous users, transparent moderation logs, an open source code base and a site that is welcoming to non-technical users. Currently, there is a single chat room for all discussions and support for user created chat rooms will be added shortly.


Suggestion: Put registration forms in the main page. Put simple captcha. Remove whatever barriers to encourage registration


> using iframes to embed comments in websites

why an iframe?


Cross origin cookie stuff can get pretty narly. Using iframes makes things easier


I spent most of the year working my ass off consulting for one of America's most hated companies building an utterly pointless system. My only consolation is knowing that I wasted a ton of their money since there's no chance it will pay any returns.


Lol. I had a good laugh.

How do you manage to continue working on something you know is a dead end though? Is the money that good? Would the money go away if you worked on another project at the same firm?


Yes, money. Also, I quit that job.


Same here! Good luck


Oh man, I don't think I've ever laughed this much on anything I've read on HN. Thanks for making my day. :)


I'm glad you enjoyed it, but it certainly wasn't funny while it was happening to me :)


Holberton School - https://www.holbertonschool.com/ a two-year alternative to college to become a Software Engineer.

The school is free to students until they find a job, then they contribute with a % of their salary. After only 9 months, many students find internships and jobs at companies like NASA, Apple, LinkedIn, Tesla, Dropbox...

It's a life-changing experience for many of our students, and it also changes the Tech industry by bringing folks with an untraditional background. Our students are straight out of high-school, some had a career before: cashier, math teacher, artist, poker player...

We have no formal teachers, no lectures, students learn by working on projects and collaborating with their peers. We are located in San Francisco and looking forward expanding.


How does this work for someone who already has a full-time job and mortgage? We would still need to take out a loan, right?


Cool that I stumbled across this, I'll be attending Holberton beginning next week.


% of their salary for how long after they have graduated?


"That is why there is no upfront cost to attend Holberton School. Once our graduates find a job, we only charge 17% of your internship earnings and 17% of your salary over 3 years."

https://www.holbertonschool.com/education#tuition


17% for three years according to their website.


Published The Tao of tmux: https://leanpub.com/the-tao-of-tmux/read. Thanks to the HN readers who bought the book!

Created new design for all my open source projects: https://www.git-pull.com (see sidebar at left, e.g. https://libtmux.git-pull.com)

Rebooted CJK (Chinese, Japanese, Korean) language project, cihai: https://cihai.git-pull.com (see also: https://unihan-etl.git-pull.com). Needs funding.

New docutils based website started, https://devel.tech. Example: https://devel.tech/features/django-vs-flask/

I catalog open source contributions I make while working on the website at https://devel.tech/site/open-source

Updates to https://www.hskflashcards.com. Switching from Bootstrap 4 to Bulma


Discovered Bulma few month ago. Like it much but have no chance to use in project. What was your point to chose it over bs?


That's a great list. How was your experience with self-publishing? Any stats you can share?


hsk flash cards seems really cool.

do people pay you for this? seems like a service and content that people would pay for, but i couldn't find any sign that this is being monetized on the site.


My health https://www.instagram.com/p/BdX6yyInrBD/?taken-by=kthakore2

Took a long break from hacking and staying indoors playing with yet another framework. I am much happier :) my depression is better and I have more balance :) Less likely to burn out.


Awesome to hear. Mental health is something that is overlooked by a lot of people in this industry until it's too late. Ran into that a few years back and like yourself, have been working to make it better.


I left an amazing company with one of the best work environments (https://webflow.com) to work for a non-profit that fights child sexual abuse (https://www.wearethorn.org). The work has been incredibly rewarding, and although I was quite nervous about leaving an awesome job and jumping into an unknown, in hindsight almost everything about the change was a meaningfully positive improvement. Working towards a mission that personally I feel has a lot of value has been an awesome experience.


I had an opportunity to work for them but declined because I didn't want to look at the content. Tell me: how much of that do you have to actually go through? And to other devs: if you have thick skin and an iron stomach, go work for them.


It’s definitely a factor to be aware of. We are legally not allowed to be exposed to image content, as developers, but personally I do have some exposure to text content as part of my role. Thorn has an excellent support / wellness system, with professional counselors, in place though for proactively staying healthy and processing what exposure we do experience.


I’m very glad they have such a good support network. A close friend of mine got a job out of law school prosecuting child sex crimes, and there was absolutely no support. After a year, he couldn’t take it any more and left public service altogether. Even a decade later, you can tell he’s still bothered by it. I have an incredible amount of respect for people who can do that work; I’m absolutely certain that I don’t have what it takes.


Definitely, I think support is crucial. I work around people who have victim ID experience (reviewing abuse material with the goal of locating children) and strongly share your respect. It’s so impactful in the lives of the victims, but you are exposed to a very dark side of humanity and are unlikely to know the eventual outcomes of the children’s lives.


No chance of contributing without a thick skin? e.g. maybe there are some infrastructure or generic dev needs as well, which do not involve working on actual "content"?


It’s a small team, so as a developer I’m not working “on content”, but definitely working “around” it if that makes sense. It’s not as bad as it could be, but I’d say somebody should at least be aware of the potential and take it seriously, for personal wellness sake.


Jacob, are you remote? This sounds like fulfilling work.


I am remote, yes! Thorn has a SF office, but the team is almost 50% remote.


I continued to work on intercooler.js:

https://github.com/LeadDyno/intercooler-js

http://intercoolerjs.org

I'm trying to get people to reconsider the more traditional web development style of server-side rendering of HTML + HATEOAS.


I remember seeing this last year too, I like it. This is a good reminder for me to try using it in side projects.

I fear though that you're swimming against the currents (obviously, speaking about the current trends of everyone using React or one of the other frameworks for frontend development). I hope you'll gain traction in creating a trend of a different path, making traditional web development style a viable alternative for modern web apps.


I don't mind going against the current, I'm a contrarian by nature. :)

I hope you find it useful, and good luck in 2018.


Yes! There’s nothing wrong with traditional web dev architecture!


Nice, I like it.


My friends have been streaming on twitch so I've been writing a bot for them to play overlays/games with their viewers, let viewers earn points, queue up music, etc.

Along the way I brushed up on some es6 concepts, learned React, and was reminded of writing eggdrop bot scripts back in the day :P

Everything is public on github and is somewhat-generic/reusable by others. Hope to complete documentation and make it 100% generic in Jan/Feb so others can use and contribute.

Core bot library: https://github.com/bdickason/hpc-bot Twitch overlay server: https://github.com/bdickason/twitch-overlay Their specific bot files: https://github.com/bdickason/dumbledore

Also helped deploy/ship Tekken Chicken, a framedata app for Tekken 7 (ios: https://itunes.apple.com/us/app/t7chicken/id1244210422?mt=8) (android: https://play.google.com/store/apps/details?id=com.t7chicken&...)


this is really cool thanks for sharing the code!


I've worked at the highest paying job i ever got in my career. But then i quit in july to focus on a startup I joined after we were featured in techcrunch and received angel investment.

Now i realized that the money was the least of our problem and the startup is on the brink of death.

I replaced the source of my income from job to stock market investments. Now i am focusing on a new side project that poped in my head. For the last 2 months I have built a prototype that works and started to dogfood it.

It's a tag that turns any object into a smart object (still working on my elevator pitch). What it does is allow you to contact the owner of any device. Put the tag on your car and anyone can contact you about your car(i.e. if it is blocking the way or you left your lights on). Put the tag on your keychain and if you lose them people can contact you. You can use the tags on anything really.

I started by building an android app but then realized you can do all this directly from the browser.

Expect the first beta in January.


> It's a tag that turns any object into a smart object (still working on my elevator pitch). What it does is allow you to contact the owner of any device. Put the tag on your car and anyone can contact you about your car(i.e. if it is blocking the way or you left your lights on). Put the tag on your keychain and if you lose them people can contact you. You can use the tags on anything really.

Did you just invent a QR code with contact details? (Edit: Not saying it's a dumb idea, sounds useful.)


Yes, most of the time qrcode is seen as a gimmick, ie you might as well write the url instead of downloading a qr reader app.

Here is my hope for making good use of it.


Can you tell me more about the problems you faced with your startup? My experience with 2017 has been the same in terms of working at the highest paying job I've had in my career. However, I AM thinking about leaving the traditional job market at some point to start my own start up in the next 3 years. I want some reality checks though.


The main problem we had is over promising. For every client we had, the CEO promised some features we cannot deliver.

Many of the financial decisions made wasted tons of money. (Microsoft Azure is not for your up and coming start up, too expensive)

I tend to favor meeting one customer at a time to understand the problem we are trying to solve, but we went for bulk email and spamming. The few clients I met one on one still believe in us today.

The startup idea was good. The solution was poorly executed, and we had too many chefs in the kitchen (5 co-founders).

I hope this makes sense.


Hah that happened at the startup I worked at. Between Sales and CEO they were promising the moon. At one point the CEO said our product used block chain in his slide deck. Not a single Dev has ever discussed block chain it was just out of nowhere.


Did they try getting bizspark? Or did they have it but overuse their limits?


We used bizspark. But that only lasts a while then you are left with a huge bill every months.


Any details on the hardware? Passive RFID, NFC, BLE or something else?


I've been playing with NFC tags, but QRcode has been working pretty well for me so far.


Interesting. Any link or place to watch the product at?


It's an before alpha product, your feedback is very welcomed.

https://www.ottomon.net


One project that I really enjoyed working on this year is https://www.meteorshowers.org/. It's a webgl visualization of NASA meteoroid data. Open source here: https://github.com/typpo/showers

I also maintain an open source SMS API called Textbelt, but it became unreliable due to spammers. I launched a paid hosted version and have been steadily improving it: https://textbelt.com/


Beautiful. Excellent work.


Lambda School - a rigorous, live online computer science education that’s free until you’re hired. https://LambdaSchool.com/computer-science

Started as a side project in January, and I had no idea it would take off like this. We now have 20 employees, including instructors from Google, Apple, Blizzard, etc. and our first graduating students are getting hired for great salaries all over the US. (Average is $85,000 in low cost of living areas)


> Started as a side project in January, and I had no idea it would take off like this

Would be interested in learning about the backstory. Did you blog about it somewhere?


Have been too busy building it.

Started with 20 students, they performed remarkably well, started adding classes each month, got into YC and raised a large seed round, solved 2 or 3 other problems that hadn’t been solved before, and now we’re ready to open the gates to more people (our acceptance rate right now is right around 2%)


Plus you sold the shit out of it. Like seriously, quality work.


It was on HN early into its life, I recall seeing it posted here. I also recall being solidly impressed by the quality of the presentation for what it was and how it worked. The value proposition clicked immediately. I think they did a pretty good job of hooking people on the concept quickly (ultimately it's pay for results, which can be a potent offer when education is increasingly expensive).


Do you accept international applicants and are the classes online or a traditional classroom setup?


Whats the deal after you are hired? a %? for how long?


17% for 2 years not exceeding 30k


And only if you’re making $50k/yr or more


Is this after tax income? Who pays the taxes?


The amount is pre tax. Students still pay taxes.


https://smart.ly

We offer a free, licensed MBA (working on accreditation process) using an interactive (re: non-video), mobile-centric content platform. In addition, we provide job-matching services for anyone interested in opting in.

I'm proud of what we've built and hoping it continues to see traction in 2018.


why was this down-voted? someone here work on a competing product in 2017 or something?


Didn't downvote you, but I would guess it's because the post sounds like an advertisement.


I left a very good SW Engineering job in London, travelled for 3 months, pimped up my 2013's side project for 3 months and started selling it. Now I have created a company around it, been profitable for months. Now I carry my boss-less/office-less job around the world as a Digital Nomad. Happy new year from Mexico!

My side project that became my job: https://readonlyrest.com


I wish such boss-less/self-employed/self-sufficient traveling was more available for developing country citizens as well (with a relatively lower rank passports).


Where did you travel? About to set off on a trip of roughly the same length in Europe. Any recommendations?


At the end of 2016 I left contracting gig for a full time position to work on helping solve intermittency problem with solar PV power generation through forecasting.

With a team of 3 including myself, the only professional software developer, we have launched and run a solar radiation and PV power forecasting/observation API (solcast.com.au) that can provide solar radiation and PV power forecasts world wide that update every 10-30 minutes based on satellite coverage.

For the past 10 months this API has been freely accessible whilst we validated our approach and expanded to cover the globe. After great feedback from users, we are now planning a big update to make it even easier to use and to integrate live PV output data into forecasting itself.

The change to work on something that contributes a large net positively to society’s around the world (making solar based electricity generation more financially attractive to operators/home owners a like all over the world) has been hugely rewarding and look forward to the growth of solar power generation in 2018.


What model do you use for power forecasting? Do you use different models for different forecast horizons e.g 1h, 1day, etc. or did you find that (recursive) multi step forecasting with the same model produces better results?


We forecast radiation just radiation data as the base for all power forecasting (conversion from solar radiation -> PV power done on the fly)

Solar radiation forecasting incorporates a few models for 1-7 days, and for now casting (0-4 hours) is based on NWP wind forecasting combined with our own cloud tracking, ML and blending with NWP models from 4-24 hours.

Some of the above might be slightly off as I’m not a meteorologist or study solar radiation modelling. Luckily Solcast founders are and we are also partnered with a project at the Australian National University working with some extremely bright people to get the science right.

If you want to get further into the details, feel free to email Nick Engerer or James Luffman (contacts on our website) about the science, or post on our community forums (forums.solcast.com.au).


That's really cool. I'd love to work on something like this.


Two machine-learning models for the detection of breast cancers from medical imaging of breast cancer biopsy tissue. Big focus on making models accessible to clinicians.

The models are:

(1) PPReCOGG, one of the models based on Gabor filters and k-NN (https://github.com/jszym/pprecogg)

(2) DeepDuct, the second model, based on a pre-trained VGG16 network and the Grad-CAM algorithm, localises lesions _and_ informs clinicians about why the model has chosen the lesion type it did.

You can find more details in my master's thesis, for which the models were written: http://cs.mcgill.ca/~jszymb/thesis/260528685_Szymborski_Jose...

(Edit: Also, if you're hiring machine learning people, medical or otherwise, please get in touch at hn at jszym point com)


Running the Cloud Native Computing Foundation, which hosts projects like Kubernetes and Prometheus, has been a very full time job, as we've expanded to include nearly every cloud company, enterprise software provider and startup in our industry. But I have gotten to contribute to a few cool open source projects:

Cloud Native Landscape (now over 350 projects and products) https://github.com/cncf/landscape#current-version

DevStats provides detailed visualizations of Kubernetes contributions https://k8s.devstats.cncf.io/

Core Infrastructure Initiative Best Practices Badge https://bestpractices.coreinfrastructure.org/


I launched two projects. 1) http://instant10-k.com/ An efficient way to search 10-k and 10-q filings for publicly listed companies. A Form 10-K is an annual report required by the U.S. Securities and Exchange Commission (SEC), that gives a comprehensive summary of a company's financial performance. 10-Q is the quarterly version. If you have ever purchased an individual stock you should read the 10-k and 10-Q reports. 2. http://datasetapi.com - A platform to host clean curated datasets.An airport dataset is live, More to follow.


Interesting, how is instant10-k different from EDGAR Fast Search[0]?

[0]: https://www.sec.gov/edgar/searchedgar/companysearch.html


Thank you! Its more efficient. Here is the sec search process 1. Visiting sec.gov 2. Clicking on Company Filings 3. Typing the ticker on Fast Search 4. Visually searching for 10-k's and 10-Q's 5. Opening in another tab. 6. Clicking on it again. 7. Finally clicking on the actual document. http://instant10-k.com takes you to the actual document instantly.


1) is so great -- thanks for making/sharing. I run into the clunky search process for 10k's once in a while and think your site does a great job improving that experience.

In case you're taking feature requests, these are some things I'd love to see in a 10k site (order in relevancy):

  autocomplete in search
  see recently released 10k's
  browse functionality
  10k specific discussions (e.g., annotating similar to RapGenius as well as longform analysis)
Some/all those may not be relevant for your site, just things I've always thought would be cool to see. And again, nice job with the site!


thank you for your feedback @liampronan! Autocomplete is in the works and I will give others consideration. Please spread the word.


Insta10K is amazing. Using EDGAR was a huge pain in the ass; this is so much easier. Thank you!


thank you! I Want to make filings accessible for all!


Instant 10k is useful and efficient, thank you


thank you!


I liked your first project's fast response time. Would be nice if you included 20-F filings as well. Otherwise from a usability standpoint, multiple sorting / grouping options for different forms would be helpful, as well as an error message for those searching for companies which don't file under 10-K (like Amdocs).


A friend and I created https://quicktype.io to generate TypeScript, Swift, Go, C#, C++, etc. from JSON sample data and GraphQL queries.

Many have tried to solve this problem – we've found at least 20 projects that attempt to turn JSON sample data into code to represent that data, but they're almost all abandoned and they all have the same fundamental flaws (they generate invalid code for most non-trivial inputs).

In the past two weeks we've created Xcode and VS Code plugins. I've had so much fun with this project! We'd love to create a business around quicktype but we haven't figured that part out yet.


What is the best C# GraphQL back end?

There is probably a small market for quickly translating SQL Server DB schema into a C# GraphQL provider with easy authentication/filtering hooks. Not sure if the market would be any bigger cross-platform since the momentum follows free open source options.


I looove it! Used it multiple times this year (Typescript / C# generator)


Woot! Glad you found it useful.

Is there anything we can improve?


I worked on two cool open source music technology projects called ListenBrainz [1] (basically an open source version of Last.FM backed by the MetaBrainz Foundation, the people behind MusicBrainz) and AcousticBrainz [2] (a project trying to crowdsource acoustic information about music and release it as public domain). We released a beta for ListenBrainz over the summer and I've been working on data dumps for both ListenBrainz and AcousticBrainz for the past few weeks.

[1] https://listenbrainz.org/

[2] https://acousticbrainz.org


Both projects look fantastic!

How does listenbrainz differ-from / compare-to Libre.fm?


We wanted to take our own shot at building a reliable open service with new tech. We're also much more focused on providing open and easy access to the data we have as conveniently as possible, so that people can build cool stuff with it. Right now, we provide a Google BigQuery dataset [1] that you can run any queries on. It has around 70M listens from 1300 users.

[1] https://blog.musicbrainz.org/2017/08/05/listenbrainz-data-is...


I got a pen plotter (AxiDraw v3) and it's been a great creative outlet. I wrote a couple tutorials on techniques I learned or discovered:

Fractal generation with L Systems: https://bitaesthetics.com/posts/fractal-generation-with-l-sy...

Surface projection: https://bitaesthetics.com/posts/surface-projection.html


This is so cool! I've been following Inconvergent's work (https://twitter.com/inconvergent). This seems like something I might be interested in.

Thanks for sharing!


In my free time I've been coding a TCP/IP stack in C++(14): https://github.com/ambrop72/aipstack . It uses a single-threaded event-driven architecture, is usable on embedded system (no mallocs), and is header-only.

Much work is yet to me done including docs (lots of Doxygen-based docs exist but introductory and TCP API docs are generally missing). However the TCP implementation should actually be pretty solid.


Built Commento, a privacy-focused alternative to Disqus: https://github.com/adtac/commento

Right now, it exists as a Github project that you can self-host, but I'll soon offer it as a paid service if you don't want to host and maintain servers on your own. (And maybe even apply to YC, who knows :))

It started out with me reading a blog post [1] and thinking "I can write Disqus tonight". And that's how it began; I had a working prototype in 24 hours (at the expense of a final exam I had in two days haha). Posted it on HN, and it blew up. And then I sat down and made it into a serious project that's now actually used by other people. I've had senior devs from huge companies (like Atlassian) contribute to the project, and I think that's amazing.

http://donw.io/post/github-comments/


There's a product named Hosted Comments mentioned in another comment on this post with a similar goal.

https://www.hostedcomments.com/

https://news.ycombinator.com/item?id=16042162


And EffectiveDiscussions, https://www.effectivediscussions.org/blog-comments (I'm developing it), it's also no-ads and no-tracking (& open source).


Yep, I am aware. In fact there's a bunch of other offerings as well (someone else with a similar product once posted a summary comment; can't find it right now).

I just quickly registered and it appears the moderation/dashboard functionality is slightly limited in HostedComments. I only have the option to view all the comments. But with SaaS commento, you'll have a full fledged dashboard. Here's a sneek peak: https://i.imgur.com/j23BfKN.png


I really enjoyed reading your code, it's got a great style. Play on :D


Thank you! :)

This was only possible because of all the contributors [1]: some of the design paradigms were entirely their ideas, so credits to them as well.

[1] https://github.com/adtac/commento/graphs/contributors


There were a few other things going on, but mostly I've taken xi editor forward. It's still almost at the point where you'd want to use it for your daily work, but not there just yet. In the last two months, I've had a strong focus on performance, and now it's paying off. A few PR's are still in flight, and there's a write-up than needs to be done, but it's now scrolling smoothly on my 165 Hz gaming monitor (2560x1440 resolution, integrated graphics at that). I'm excited about the progress and feel that it will become a really usable tool fairly soon.


I've worked almost exclusively on https://info-beamer.com, a digital signage service for the Raspberry Pi. It all started as a for fun project while freelancing and turned into a profitable business. All while still being a lot of fun to work on. Unlike other Pi based solutions, info-beamer isn't using a sluggish browser but uses Lua/C at its core to use all the hardware acceleration features while still being pretty simple to program (see https://info-beamer.com/doc/info-beamer). The challenge is always to make all those feature easily available through a web interface. I'm currently improving that a lot by enabling users to create more complicated output without any programming knowledge. Getting to know Vue.js has helped a lot with that.


I was working on LibreTaxi https://github.com/ro31337/libretaxi

it's ride sharing app that works thru Telegram (currently). Surprisingly, it worked really well, there are 100-500 rides in some cities every day


I was excited about this project when i first saw it here but i have to say the setup instructions are very confusing. It would be nice if the documentation had step by step installation and first boot instructions. Many months after cloning the repo, i have never managed to set it up.

The instructions are all over the place. The NodeJS instructions say you can run npm -i after installing node. Then you get to Getting Started and it seems like a detached process altogether. The getting started section should start from either installing the dependencies or cloning the repo. Right now it just starts with renaming some file.

Good work, though, at least from what i can see on the site once it actually does run it is a beautiful thing.


Thanks, I know, some setup is required. I ask everyone who had any issues to update instructions in confusing places. So you may want to check it again.

Also, I'm going to add docker image soon, so it will just work with few commands


At work I've got more familiar with Terraform, got started with Kubernetes, and contributed significantly to the infrastructure of a few awesome projects.

In my spare time I've been maintaining my Autospotting pet project, which is maturing nicely, growing a lot and already generated savings in the six-seven digits for its users: https://github.com/cristim/autospotting

I also spent time learning to play guitar, made a habit of practicing and working out on a daily basis and towards the end of the year I became a father.

All in all it was likely my best year so far.


I've been working with Terraform and Kubernetes a bit as well. Their strengths seem to be very complimentary when aiming for a platform agnostic system.


I worked on building a python library for automated feature engineering called Featuretools (https://github.com/featuretools/featuretools/). I had been working on it for 2 years, but in 2017 we separated it from the rest of the codebase and made it open source.

Even though feature engineering is crucial for building machine learning pipelines, there are few formal methods for performing feature engineering. We see Featuretools filling a missing component in the software engineering stack for data science.

It has already been put to the test with our customers at my company, but we have also begun to release demos so that others can pick it up https://www.featuretools.com/demos.


I nearly ended up a pancake when my steering failed heading into a turn in February, the day before my birthday. I realized the circumstances that led to me not being able to fix it before it led to near disaster weren't going to change on the path I was heading down largely by habit, so I finally set some priorities.

I decided to focus on making a business out of music. I'm far from where I want to be, but it's been a long time since I was doing Mechanical Turk tasks to pay for junk food. I have savings, my music is improving, and 4 people pay me almost $15 a month through Patreon[1]. Probably not a lot to the crowd here at HN, but it's a peace of mind I never knew before.

The big, super-important lesson I got from that is to not cling to what I wanted at some point in the past and accept how things are. I wanted to be fully financially independent, but had no plans, no goals, no notion of how I might make it happen. I had the desire, but not the will or commitment.

Being two seconds and one failure of attention from the front end of an 18-wheeler has a way of hitting the reset button.

[1] https://www.patreon.com/digitalscofflaw


Music is sooooo hard to make a living at. I burned through my savings and am getting back into engineering. I love music, but it won’t feed 4 kids, even if the music is stellar. Cheers to you!


I'm more on the tools and services side. I make melodies, synth presets, sound effects, and stuff like that. Still hard, but less of a gamble. I tried the other side before, and you're right. I just barely made enough to pay for the headphones I use for mixing.


2017 has been an intense, yet extremely rewarding year to say the least. This year I learned the meaning of "grit".

- Launched my startup on Product Hunt (https://www.producthunt.com/posts/slackpass-2).

- Interviewed with YC, sadly didn't make the cut.

- Had hundreds of of calls and thousands of chats with founders looking to create paid communities.

- Helped many create their own profitable, paid communities.

- Became a solo founder.

- Became profitable enough to cover both business and personal expenses.

- Rebranded to LaunchPass (https://launchpass.com) due to inevitable trademark issues with the use of "Slack" in our name, and plans to expand beyond Slack. (btw Slack has been awesome regarding the transition)

- And plenty more I intend to write about in a "year in review" post I'm working on.

Becoming a founder this year was one of the most challenging, fascinating, and deeply rewarding experiences I've ever had.

Here's to a happy, healthy, productive, and successful 2018

Happy New Year HN!


Started working on my first side project! It's an alternative way to screen developers that I think is better than anything that's currently out there. I'm interested in hearing feedback on the idea.

You provide JSON data that will be exposed through an API which candidates will use. They are given instructions on how to parse and manipulate the data. Then they POST the response to you. If the response is 200 OK - they've passed and they can upload their code for your team to review and decide if they should go to the interview stage.

I think this has lots of benefits:

- It's gives candidates a real-life problem to solve. Most, if not all software developers will have to interact with API's and manipulate data.

- Candidates can use their own dev environment that they are comfortable using.

- It saves the company time. They can choose to only assess the code of people who pass the test.

- It makes for a good candidate experience. I think it reflects well on a company if their interview process is close to real-life work.

Hoping to ship the beta version of this next month


This approach shifts the effort on constantly changing the source JSON data and specification on how it needs to finally look to the client. Lazy devs will inevitably post solutions online.

This approach also requires the client to implement ratelimiting, but you could fix that by having the data POSTed to your servers instead.

I unfortunately can't remember the name of it, but a service was presented on here a while ago that presented a series of security challenges then connected successful applicants with employers and managed the whole process. Methinks that's the right way to do this kind of thing.


You're thinking of Matasano?


Taking my company's first product from "demo to industry partner" to "two systems installed and used in production by our end clients".

It's been a rough year financially but we've made a ton of ground and it's looking pretty damn shiny for 2018.

Edit: Since that was pretty vague, it's a system for guarding, automating and remotely operating industrial hydraulic booms (eg. fixed plant rockbreakers, jib and knuckleboom cranes, etc.)


Worked on a dating app http://crushhourapp.com as a side project with a friend of mine.

It took almost 1.5 years to complete, Backend APIs were done with Django, iOS app with Swift. The concept is dating app for London commuters.

I created the whole London Underground maps programmatically in the app. The final result was ok, unfortunately dating market is already saturated, and our market is only limited to London, lesson learned, test your idea first, build a quick prototype, don't spend more than 5-6 months, unless you are really sure.

For the rest of the year, I have concentrated in learning Reactive functional programming, created a small backend app with Clojure, at work I am working on iOS app, which I have architected using, RxSwift, MVVM, it has over 650 tests, with close to 80% test coverage using Quick and Nimble frameworks.


How do you feel about RxSwift and MVVM so far? Do you see it becoming increasingly more widespread in production code?


I think RxSwift and MVVM architecture work great together. I know some large companies that would keep it simple, would not use any framework. Some are stuck with legacy code, that requires lots of effort to change, but most of the time you will be asked about different architectures during an interview.

I would say it does not really what architecture is used, as long as it is easy to test, simple to add new features, any architecture will be fine.

From my point of view, I probably have a bit different approach to MVVM where we have Service layer (similar to Java Spring), below View Model, stick to the simple rules like single responsibility, separation of concerns, define your rules about communication between the architecture layers. RxSwift makes the communication between the layers so much easier, the old alternative would be using delegation pattern, notifications. In a large project over 30k lines delegation pattern will become painful to manage. Choose the right architecture for your project and don't be scared to add additional mechanisms, layers, because there is no solution that fits all.

With RxSwift concurrency is handled for you, you can do the heavy duty tasks in the background, observe on the main thread for ui updates. You can run multiple tasks in parallel and return the combined result. Things like these, become trivial to work with. Another benefit is that you will strive to write more functional code, functions that return observable that does one thing, and chain other observables, It wont be pure functional since most of the you will have side effects, which is fine. But still it is a good fight between the Object oriented paradigm vs Functional programming paradigm, finally you will settle with both and find a good balance.


I started a company, Argonomo (https://argonomo.com) and we founded and soft launched two ventures:

- SafeWhistle: An anonymous, encrypted, privacy-focused whistleblowing and incident management application companies and institutions can implement to help cut down on lack of reporting and increase transparency. (https://safewhistle.com)

- Sidepitch: A venture management system targeting private equity groups and venture capitalists. Streamlining the application process for startups and giving investors a central management solution for their investments, instead of a collection of emails, paper documents, and in-face communications. (https://sidepitch.com)


Built a fintech-ish startup all by myself, zero funding. I didn't intend to do it all by myself but it's not a sexy business so no one was jumping to go full-time. It came close to breaking me as a human being, honestly.

Anyway, I went from being a CTO who was constantly being pitched horrible no-good business ideas by first times CEOs - who as a rule, wanted to give me 10% equity but also wanted me to build the project for free - to a CEO who closed countless sales and knows his CAC and LTV like the back of his hand.

If I can, so can you but you have to manage CAC:LTV.


Do you have a link where we can learn more about you / your startup?


My open source project: building best practiced apis fast with Python3 https://github.com/agconti/cookiecutter-django-rest


I released another iOS app. It has been two years since I released anything. Really lost interest because the market is so large, and it’s hard for a ream of one. However, I’ve decided to write several little apps instead of trying to boil the oceans.

https://h4labs.wordpress.com/2017/12/29/word-search-1-1-rele...

The idea is that quantity trumps quality:

https://blog.codinghorror.com/quantity-always-trumps-quality...


I left my full time job about 3 months ago to start my own software consulting venture. Ive maintained projects from old clients I had on the side, and also created and launched a new project for a client (RN mobile app). Still have to get an online presence setup for the company. Its been going well so far, and I am looking forward to new projects and clients in 2018.

I launched ScrumGenius (https://scrumgenius.com). Its a side project I started for fun at my previous job (it was just a simple slack bot script back then) and decided to actually build a service and launch it a few months ago. I did not take it too seriously at first, I was just using it to learn. However, after reading indiehackers and other people launching products I was really inspired to give it a try. Its been steadily growing and it makes around $300/mo.

Hoping to continue to grow it even more in 2018.

If anyone is looking for a end to end consultant that does Full Stack Dev with experience in mobile and web! Please do reach out, would love to talk! I am based in Canada and UK.


A friend and I were concerned that Vimperator would die with Firefox 57, so we made our own version [1]. To our delight, it mostly works and has a few users.

[1]: https://github.com/cmcaine/tridactyl


This has allowed me to keep my sanity (and for the most part keybindings) in the newer FF releases; many thanks!


I put time into my search engine for talks (https://www.findlectures.com/). I did two conference talks on it, at Solr/Lucene Revolution (https://www.youtube.com/watch?v=e_ia1DRz3l8) and AI With The Best (https://www.youtube.com/watch?v=dvUf9LZxEv8).

I've also been working on extracting the UI components into a library of UI components for Solr (https://github.com/garysieling/solrkit) and a project to generate email alerts of suggested talks based on interests (https://github.com/garysieling/email-alerts).


I started streaming coding on Twitch - https://twitch.tv/mimmingcodes

It’s rekindled my excitement of using the Internet to share knowledge.


I've never watched anything on twitch, but I just sat through a bunch of your old recordings, and really enjoyed it! I'm a dev and really like pairing, where I work now they hate it, this was kinda nice to relive that pairing-ish vibe :D good work :D


Automating more Terraform stuff.

First, a script that calculates what percentage of your AWS resources (15 different resources for now) are managed by the Terraform code in a given directory, and then creates GitHub style badges for each. https://github.com/chrisanthropic/terraform-infra-as-code-co...

Second, a script to fully automate importing an existing GitHub org into Terraform and create a basic Terraform resource block for each resource. Imports teams, users, user memberships, and all repos. https://github.com/chrisanthropic/terraform-import-github-or...

Both scripts are just bash and the AWS API, GitHub API, and Terraform. jq is also required.


https://ossia.io : a visual programming language for interactive shows & music, and its associated network protocols & integrations in creative coding environments (puredata, max/msp, unity3d, openframeworks, etc...)


1. on-device speech recognition and command clasifier. V. Proprietary.

2. antigen target filtering system for a boolean logic platform for using CAR-T with AML.

3. Database and retrieval system for a series of experiments in gerbil and chinchilla cochlea to study wave propagation along the organ of corti.

4. a unity-based traveller RPG character management suite.

5. A system to measure whisker deflection in rats as a proxy for studying Bell's palsy

6. a variety of small silly projects for personal use.


links/info for 2&3 please! Sounds really interesting.


2. https://github.com/gvoysey/tetrad (in development)

3. code forthcoming after publication :)

mostly, this is work done at BU and it’s spinoffs.


Thanks!


where are you doing this ?


CPU INFOrmation library: a cross-platform library to discover supported instruction sets, microarchitecture, and cache parameters of the CPU. Started as a "oh, I can do it over the weekend" project at first, took close to a year to get to production quality.

https://github.com/Maratyszcza/cpuinfo


I spent some of my time creating realtime procedural animations in 64kB, as part of the demoscene. 64kB is the size of the Windows executable (including all models, textures, music, etc.).

Youtube capture of my last work: https://www.youtube.com/watch?v=27PN1SsXbjM (but please try the executable if you can)


Started my own company as a side gig and started selling my own software. I developed PowerShell Pro Tools for Visual Studio as well as a web site development kit with PowerShell. https://ironmansoftware.com/ It's been super rewarding and a lot of work. Very excited to see how it turns out in 2018.


I open sourced a ZIP library written in Swift for macOS/iOS/tvOS/watchOS and Linux: https://github.com/weichsel/ZIPFoundation and wrote an article about it here: https://thomas.zoechling.me/journal/

Also improved my Mac app to record and export Animated GIFs: https://itunes.apple.com/app/claquette-animated-screenshots/...


I left my old job and built a lot of things, even built some things with friends. A lot of which have become open source. Also did a lot of reverse engineering.

1. https://labs.maplestory.io

2. https://maplestory.design

3. https://maplestory.wiki

4. https://github.com/Inumedia/NXLDownloader

5. https://labs.crr.io/maplestory/PKG1

Those are the main ones.


This is cool! I didn't realize people were still playing maple story :O


As a side project, I worked on https://stayintech.com/ It was fun to explore the Google Maps and Places API.

And also I updated my usability checklist https://stayintech.com/UX


After my former employer had a successful exit, I spent a few months helping to integrate with the new parent company. Once that was done to satisfaction, though, all but four members of our tech department were laid off--including me. I used the bonus payout, my severance and some investment success to finance my own startup...

My first product is meant to help businesses with eCommerce stores (particularly those powered by WooCommerce for now) keep track of inventory counts and locations:

https://stime.tech/yoink/


Quit my job (Frontend engineer), moved to the east coast and bought a hotel. I don't think I can go back working for corporation/startup.

My net income reduced (Based on the 2016 P&L) but so far I like it.


Learned about Go and blockchains by combining them in a personal project. For those already seasoned in Go or Blockchains, feedback welcome :) https://github.com/Grrrben/gocoin


I focused on completing my Master's in Biomedical Engineering. I looked at augmented reality for guidance during a surgery itself. The clinical application was the (robot-assisted) laparoscopic partial nephrectomy, aka kidney cancer surgery. Knowing that surgeons use ultrasound imaging during the surgery to scan the kidney, I sought to answer the question of how can we leverage this information to guide the surgeon and inform them of where their tools were in relation to the tumour at any given time?

A relatively easy to read description can be found in [0], while the main paper can be found in [1].

[0] http://stories.innovation.ubc.ca/augmented-reality-in-minima...

[1] Singla, Rohit, et al. "Intra-operative ultrasound-based augmented reality guidance for laparoscopic surgery." Healthcare technology letters 4.5 (2017): 204. http://digital-library.theiet.org/content/journals/10.1049/h...

More

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: