if a candidate didnt get through the traditional method (including multiple interviews and code assessments) then maybe they're just not the right fit for that specific job...why prolong the process even further?
Launch HN: Hatchways (YC S19) – Internships Instead of Interviews
51–60 of 83 posts
Re: Launch HN: Hatchways (YC S19) – Internships Instead of Interviews
#52if a candidate didnt get through the traditional method (including multiple interviews and code assessments) then maybe they're just not the right fit for that specific job...why prolong the process even further?
Hatchways thesis is that the traditional method (multiple interviews and code assessments) sucks for screening talent.
Here's an analogy. The current method is like qualifying for a marathon by having candidates run a 100m sprint. A better method is using a half-marathon as the qualifying test. Great marathon runners will probably pass both the 100m sprint and half-marathon tests. That said, some pretty good runners will fail the 100m sprint but pass the half-marathon test with flying colors.
Hatchways is changing the qualifying criteria from 100m sprint to a half-marathon so more people can qualify for the race!
Re: Launch HN: Hatchways (YC S19) – Internships Instead of Interviews
#53This is great - what you guys are doing particularly hits home for me. I graduated out of UC Riverside, with a degree in Accounting, and if any of you are Asian-American in California, you'd know it's infamous nickname: University of Chinese Rejects. I'm willing to bet that my GPA was even lower than yours Jaclyn. I managed to finagle my way into a UX/UI position at a major consulting firm and before that STRUGGLED t…
Thanks for sharing your story. Glad to hear you were able to overcome the traditional barriers. Great idea to explore design as an industry we can expand into in the future!
Re: Launch HN: Hatchways (YC S19) – Internships Instead of Interviews
#54I seriously hope you folks succeed. All the luck to you! Here's my non-traditional story: I've worked on my own personal projects for 10 years, always looking for new challenges. I've dropped out of college after 2 years since it was crystal clear to me that I learned much more when working on my own stuff. I would learn new things in college, but at a much slower pace. I guess it was useful to learn calculus... Anyw…
In my experience, you don't need to be a major contributor to an open source project to gain notice. I've only 3-5 minor contributions (work rules generally preclude me from being more active)
While my contributions haven't been large in terms of LOC, nor have I added new features, I have identified, debugged and supplied patches for bugs.
I mean, I did this out of self interest, of course.
A defect in a certain DB lib on bulk copying with nulls would fail (uninitialized memory in the C portion of the lib). That took about 6 weeks of on again/off again attention. Debugging a C module in Python is a bit of a pain in the ass.
Fixing 64-bit compatibility in a Windows ODBC C++ wrapper was another. This was a pain because it required a line by line audit to make sure the correct types were used on every ODBC call.
A problem with some inline assembly using rtdsc as a highres timer reporting 200 years had elapsed in a few minutes. This took a long while to figure out and only stumbled into the cause of the bug. I was using the timer to measure socket timeouts. We were supposed to timeout after 30 seconds, but randomly, the timeout would trigger immediately. Stumbled into it because we were using the same timer to measure database queries. Found it odd that my logs were reporting queries were taking more than 200 years to complete, as I sat there reading the logs and seeing them complete.
Another was an issue of file handles being, no pun intended, mishandled on a .Net logging lib. I forget the details on this one. Dont think the file was being properly closed at exit?
Reported an issue with Boost DateTime overflowing on certain arithmetic situations. Coworker found the issue, but I narrowed down the issue, figured out a solution and reported upstream. Finding the wrong typeset in a template zoo like Boost is an arduous task.
I'm sure there's more I've forgotten, but my contributions haven't been much in the way of LOC. My contributions have been in the hours spent and expertise in finding the typically 1 or 2 lines that are wrong.
Having experience like that on a resume (and being able to talk knowledgeably about what you did and how you do it) do open doors. Having "found and fixed obscure bug" in a widely used OSS lib is nearly just as valuable as churning out lots of OSS code. Probably more noticeable if you fix bugs in an OSS project that lots of people/companies use, rather than churning out tons of lines in a project almost no one has heard of or uses (beyond the authors).
Edit: spelling
Addendum: providing bug fixes for 3rd party libs also demonstrates an ability to work with and understand code written by someone else, which is a valuable skill in of itself.
Re: Launch HN: Hatchways (YC S19) – Internships Instead of Interviews
#55I really wish you all the best. A large part of our team at RedCarpet was hired this way. The biggest thing that it helped was to not depend on signalling from the big name universities in India (the IIT, etc).
We are proud that we did this and is now part of our company culture. The internship program is core to our people ops and everyone in the company participates.
We call it the ROCKS program - https://www.redcarpetup.com/rocks/
Today our success rate is 100% placement - either at RedCarpet or other startups in India. From what i hear, our internship program is now used as signalling by others (without depending on college name branding)
Joke : we internally describe the ROCKS program as "its like Lambda School, except people get paid to learn and get a job". Feel free to steal that ;)
Re: Launch HN: Hatchways (YC S19) – Internships Instead of Interviews
#56if a candidate didnt get through the traditional method (including multiple interviews and code assessments) then maybe they're just not the right fit for that specific job...why prolong the process even further?
I would guess in most cases candidates can't fully be their authentic self or perform at their best during the interview process, so for me I'm not sure I can boil it down to "fit".
Our hope is that we can remove the traditional method so candidates don't need to go through a lengthy process as a whole, and instead get on the job as quickly as possible via one interview!
Re: Launch HN: Hatchways (YC S19) – Internships Instead of Interviews
#57if a candidate didnt get through the traditional method (including multiple interviews and code assessments) then maybe they're just not the right fit for that specific job...why prolong the process even further?
Hatchways thesis is that the traditional method (multiple interviews and code assessments) sucks for screening talent. Here's an analogy. The current method is like qualifying for a marathon by having candidates run a 100m sprint. A better method is using a half-marathon as the qualifying test. Great marathon runners will probably pass both the 100m sprint and half-marathon tests. That said, some pretty good runners…
Love this analogy. Everyone showcases their skills in different ways, and we definitely want to help more people shine in whatever run they perform best in.
Re: Launch HN: Hatchways (YC S19) – Internships Instead of Interviews
#58I really wish you all the best. A large part of our team at RedCarpet was hired this way. The biggest thing that it helped was to not depend on signalling from the big name universities in India (the IIT, etc). We are proud that we did this and is now part of our company culture. The internship program is core to our people ops and everyone in the company participates. We call it the ROCKS program - https://www.redca…
Appreciate your comment! Will look more in-depth and perhaps reach out in the near future to pick your brain further (probably after the yc demo day rush though :P)
Great website and mission. Wishing you all the best too!
Re: Launch HN: Hatchways (YC S19) – Internships Instead of Interviews
#59"Mainstream hiring processes are biased towards those who went to great schools, had high GPAs, are native-English speaking, have worked at brand name companies, are extroverts and great networkers, etc. But there are so many talented people who don't fit that profile. We're excited to work on ways of hiring that give them a better chance." This really struck a chord. All the best!
Re: Launch HN: Hatchways (YC S19) – Internships Instead of Interviews
#60I would much rather do a day of work than jump through abstract technical tests and non real problems.
You could add a point for each of:
1) quick setup
2) selected a bug and started working on it
3) wrote a test or two
4) solved the bug
Extra points:
5) made the codebase better
6) asked questions
7) etc.