At every company I've been part of the interview design process, I always insist on having practical tasks with real tools because how quickly someone can parse documentation and code context are not trivial aspects of the job. It does take a lot of time to set up sandbox projects however which a platform like this does away with. Looking forward to the day when no candidate sees Leetcode or HackerRank as part of a tech interview again.
Thank you, and love that you promote great interviewing practices! What were some examples of prompts you routinely used for interviewing? And any learnings on great (or not so great) interview design?
This was my process: built a sandbox/simplified version of the app and hosted on a separate repo. Told the candidate to either set up their laptop with the libraries installed (friction) or to come onsite an use one of out computers (not possible in remote any longer). Built a 2-3 hour onsite task plus a half-day take home extension that were features already in the production app. Did this for hiring iOS, Unity, React and Python devs and it worked pretty well for all these roles.
When academic teams collaborate on a paper or a poster, what do you think would be more useful- real-time WYSIWYG editor or "here's the file, make the changes you want and then let me know"? Github/Invision/Figma or file servers?
Having worked as a researcher, I can tell you it's the latter.
Because these teams have never used (eg) google slides to put together a presentation from 3 different continents
I have, and I’ll never go back to PowerPoint. Not coincidentally, my lab uses biorender all the time and it saves us an absurd amount of time. We often paste directly from biorender images into google slides. Yeah it’s pedestrian, but it works really, really well.
Ask the penny pinchers who look down their nose at this who funds them, how many R01s/R37s/equivalents they hold, etc. $400 for a decently run lab is jack shit. Unless you’re buying Rain-X by the gallon at Sam’s Club and silanizing your own float glass for gels, I suspect there are other expenditures that offer less marginal value. YMMV, but it probably won’t.
These guys are doing well because they solved a long standing problem that is worth a lot to a large group of people to make it go away. Every single signup I’ve seen went like this:
“How’d you put that together in 5 minutes?”
“We used biorender.”
“What’s biorender?”
“A web application that helps put biomedical illustrations together. If something is missing, you ask and they add it.”
Part of the concern though is that should biorender shut down, for whatever reason (or some other human factor comes into play), you may lose access to anything stored there. You are also entirely dependent on access to that service. This is in contrast to a desktop application - YOU control the data and access to the service. It would be completely doable to have collaborative features built into that too without needing a third party to be involved, just not as profitable.
It's bad enough that we have to deal with Elseveir - we really should not be encouraging stuff like this when it is not necessary as a general principle.
I download the illustrations (duh?). That’s the value. I don’t actually care about the service per se. Once the illustrations are in the paper/talk, the value has been realized, for my lab.
Nothing in this world lasts forever. I’m ok with that. Software undergoes bit rot, and services shut down. For us, the value is simply that we can communicate complicated experimental designs and results, clearly and effectively, without a lot of application training or other bullshit. That’s worth a LOT to my lab.
We release almost everything open source, and are militant open data proponents. We also like to remind people “if it breaks, you get to keep the pieces”. The value of this service isn’t in the pieces.
The quoted number is actually the entire budget, including marketing.
> Media analyst Arvind Bhatia estimated the game's development budget exceeded US$137 million,[7] and The Scotsman reporter Marty McLaughlin estimated that the combined development and marketing efforts exceeded GB£170 million (US$265 million), making it the most expensive video game ever made at its time
That's not entirely true. They try and imitate your bank's branding on the log in page and do not make any mention of Plaid. For example, when setting up Venmo, I thought I was logging into something my bank had created.
I mean, the only reason I even know what Plaid is is because the services I've used advertise they are using Plaid, for example, Drop: https://imgur.com/a/l4PM6QG I remember seeing it on Citibank too.
You're still sharing your bank account information with someone else. Even if it's your bank's API or whatever, "something my bank created" could be "something my bank had hired an external company to create," or even "a front end my bank created that uses third party software to do all the data processing on the back end." I'm not sure of a meaningful distinction between each case. If you want to minimize sharing bank account information "for privacy" then you don't give your bank account information to anyone.
> If you want to minimize sharing bank account information "for privacy" then you don't give your bank account information to anyone.
That's the whole point. You don't know you're giving your account information to anyone. I use Venmo and had no idea they relied on this technique until reading your comment.
Is the gist of this company logging into a bank's web service using a user's credentials and scraping their account data and exposing that data via APIs to other developers?
I thought they actually integrated with the banks on the backend, but if this is all they do, I'm not comfortable using any product that snoops my bank info without any accountability.
