Enveritas is a 501(c)3 non-profit working on sustainability issues facing coffee farmers around the globe. We provide sustainability assurance for the coffee industry. We visit smallholder coffee farms around the world to understand their social, economic, and environmental practices. In 2025, we will visit over 100,000 farms across more than 25 countries in Asia, Africa, and Latin America. We work with leading coffee roasters to understand the sustainability issues in their supply chain based on our sustainability standards.
Enveritas is a 501(c)3 non-profit working on sustainability issues facing coffee farmers around the globe. We provide sustainability assurance for the coffee industry. We visit smallholder coffee farms around the world to understand their social, economic, and environmental practices. In 2025, we will visit over 100,000 farms across more than 25 countries in Asia, Africa, and Latin America. We work with leading coffee roasters to understand the sustainability issues in their supply chain based on our sustainability standards.
Enveritas is a 501(c)3 non-profit working on sustainability issues facing coffee farmers around the globe. We provide sustainability assurance for the coffee industry. We visit smallholder coffee farms around the world to understand their social, economic, and environmental practices. In 2025, we will visit over 100,000 farms across more than 25 countries in Asia, Africa, and Latin America. We work with leading coffee roasters to understand the sustainability issues in their supply chain based on our sustainability standards.
The gold standard for this in sensory analysis is a triangle test — which I happen to have done with coffee from Ninth St Espresso, who sells a regular and substantial identical decaf. We brewed 3 identical batches (where 2 were the same beans and the 3rd was the other bean). In an office with ~12 tasters, the ability to pick out the “different” beans was 33%… ie random chance.
If the above sounds confusing, consider red wine vs white wine… visual inspection alone would get you 100% accuracy.
I used to believe decaf processing would have to change the taste, but empirically, with admittedly untrained tasters (but ones who know coffee very well), we couldn’t tell.
Decaf is simple to pick out by a person who is competent at tasting coffee. As easy as your red/white wine visual test. People in general are very bad at tasting and especially thinking and communicating about tasting. Plus also people may not know what decaf coffee tastes like or may have never thought about it before.
Speaking from my own experience, which may be different from the grandparent comment: I’ll ask ChatGPT (on GPT4) for some analysis or factual type lookup, and I’ll get back a kinda generic answer that doesn’t answer the question. If I then prompt it again, aka a “please look it up” type message, the next reply will have the results I would have initially expected.
It makes me wonder if OpenAI has been tuning it to not do web queries below some certain threshold of “likely to help improve reply.”
I’d say ChatGPT’s replies have also gotten slowly worse with each passing month. I suspect as they try to tune it for bad outcomes, they’re inadvertently also chopping out the high points.
> touchid sometimes fail due to grubby fingers. Can't malware spoof a "can't read fingerprint, please enter password" dialog?
That would require the malware to be able to determine the timing of when your finger pressed the TouchID sensor, which I suspect is not accessible above the OS layer.
TouchID is a great solution for this. However, the root issue remains: social engineering the user to allow admin privileges when not necessary… there are still too many cases of requesting elevated privileges. Maybe signed software with entitlements can sufficiently solve that? But I’ve seen way too many users who “trust” email attachments or phishing emails…
I worked at WebTV before it was public… back when it was called Artemis Research and had a website that proclaimed we did research in sleep deprivation of rabbits (there was a pet bunny rabbit that wandered the office, occasionally making a mess of things).
I love how wonderfully weird the web was back then! Such a different world; so hard to explain today to those who didn’t know it. And an incredibly talented team.
WebTV became the foundation for Microsofts hardware biz, and is responsible for the Xbox. If you go back farther. Artemis was responible for the Xband modem and is largely repsonsible for online gaming as a whole.
We finished a hiring process about 2 months ago, along with two other hires earlier in the year. As a hiring manager, 2023 has been brutal — 1,000:1 application to hire ratio (based on stats in Lever; no exaggeration).
I’ve made a point to try to be as open and as fair as possible in our hiring process. We list the salaries up-front in the job description. We create per-hire ”join our team” pages that share a lot more about the role, including the exact hiring process, links to docs with interview prompts, timelines, and who you’ll interview with. We don’t require any formal studies; we hire globally through an EOR; we don’t change salary based on where in the world. We’re also a non-profit with a compelling mission and interesting technical challenges. We want to hire people who bring different viewpoints and add stuff to our team, and where we can give them a good professional experience too.
A few observations on folks who are applying:
- About a third are just outright unqualified. For example, one of the roles needs folks with experience in Postgres at a medium level (triggers, plpgsql, replication, PostGIS, etc) and we’ll get applicants who’ve only used an ORM to work with Postgres). We have a few screening questions that literally confirm the required skills listed in the job posting and use these to auto-reject applicants. (Again, nothing unfair or tricky; just literally “have you used features like triggers or replication in Postgres?”)
- About half seem like good candidates from application but are obvious-no’s after either several-minute examination of application materials or a 20-30 minute call. (Generally failing a screening call because they’ve exaggerated on their resumes; haven’t read our “join our team” page that we ask them to read before our call; have red-flags in our call; etc.)
- The remaining ~15% are reasonable folks for us to do technical interviews, and it comes down to how much their experience lines up with the areas we need and how well we can asses their skills in interviews. (Two-way street, of course… lots of chances for them to ask us Qs!)
Where it’s been tough is filtering through the top of the funnel. In part: ChatGPT has really made a difference, in that many candidates are now using it along with much more sophisticated tools to track all their potential jobs. I think it’s _good_ in some ways, but previously we could use the ability to write well (resume, answering Qs like “why do you want to work at our company?”, communication in email) as a good proxy signal of overall effectiveness of communication. (For a fully distributed remote team that does its work via slack and GitHub, this is a relevant skill.) So we’re now having to do a _lot_ of extra work to try to be fair to everyone and keep bias as much as possible down. (I had to hire a contract recruiter to work with us - it used to be resume screening and initial calls were an average of 45 minutes a day; it went to 3 1/2 hours this year; as a CTO there’s no way I can spend that kind of time).
I don’t know what the solution is — it feels like a bit of an arms race. For every company that’s trying to run a good process, the extra application load is a real cost; for companies that aren’t particularly thoughtful, it makes it worse for the candidates.
I’d love to hear what others think about anything I’ve shared. What can we as hiring managers do to make it easier for you? And: what can we do to make it easier on us?
This mostly sounds good but I wonder if you'd make it better by dropping the strict postgres requirements.
I've never used postgres. I've worked with umpteen technologies over the years, including lots of gis, just never needed to use postgres yet. But I know it's out there and I'm confident if I needed to, it wouldn't take more than a few days to catch up.
Not a dig at you specifically but a lot of companies seem to look for specific tech experience rather than a track record of just learning whatever tech is needed.
I hear you! We’ve definitely had candidates we’ve talked with in the past who didn’t have experience with the tech stacks involved. In our case, we’re not suffering from a lack of qualified candidates (quite the opposite!), so we can afford to be more strict in requirements.
I remember a friend describing interviewing as a search problem. How long and how much resource is one willing to spend on a search? And then, adjust the hiring process (for employer) or job search process (for candidate) accordingly.
Enveritas is a 501(c)3 non-profit working on sustainability issues facing coffee farmers around the globe. We visit coffee farms to understand their social, economic, and environmental practices. In 2023, we will visit over 50,000 farms across more than 20 countries in Asia, Africa, and Latin America.
Enveritas is a 501(c)3 non-profit working on sustainability issues facing coffee farmers around the globe. We provide sustainability assurance for the coffee industry. We visit smallholder coffee farms around the world to understand their social, economic, and environmental practices. In 2025, we will visit over 100,000 farms across more than 25 countries in Asia, Africa, and Latin America. We work with leading coffee roasters to understand the sustainability issues in their supply chain based on our sustainability standards.
* Backend Software Engineer - $140-$155k — https://enveritas.org/jobs/backend-software-eng/#10d7adef8us (worldwide remote)
* UX Product Designer / Engineer - $130-$150k — https://enveritas.org/jobs/product-designer/#10d7adef8us (worldwide remote)