cporios

Schema Labs (OpenCouncil) | Civic Tech Software Engineer | Athens, Greece | Hybrid (3 days in office) | €55K-65K

We're a 4-person non-profit building OpenCouncil (https://opencouncil.gr), an AI platform that makes Greek municipal council meetings accessible to citizens. We take hours-long council sessions and turn them into searchable, per-topic video clips, SMS/email notifications, and speaker analytics.

We're live with 10 municipalities including Athens, and the project is open source: https://github.com/schemalabz/opencouncil

Stack: TypeScript, Next.js, React, PostgreSQL, Elasticsearch. Heavy use of LLMs for transcription, speaker ID, and turning unstructured PDFs/transcripts into structured data. We use Claude Code, Cursor, and similar tools daily.

Looking for: 2+ years experience, strong web dev skills, genuine interest in civic tech. Bonus if you're strong in either data engineering/LLM integration or frontend/UX. Must be based in (or willing to relocate to) Athens, fluent in Greek and English.

We offer up to 30% annual profit sharing, lunch on office days, and up to 3 days/week remote. Small team, high ownership, real impact.

Details & apply: https://schemalabs.gr/jobs/civic-tech-software-engineer

if anyone's wondering, the post title is wrong -- both of the first two characters are en dashes (U+2013).

Exactly, which is the entire reason the array pattern exists, and yet the article forgets to mention that entirely.

We don't want to replace all interviewing. We won't test for culture fit, or presentation skills (although we can test for some industry-specific knowledge and communication skills). You can think of it as technical screening stage.

These are all great questions, thank you.

We don't want to serve non-hardware companies that are hiring their first hardware engineers. We want to serve companies of all sizes (including startups) that are building products that are so critical that they can't waste their senior engineers' time to do 10-20 technical interviews a week. These customers will care a lot about their hardware engineers.

Yes, our customers should do more interviewing for candidates that pass the Touvlo interview. We think that our service should be primarily a technical screening before the company invests more in additional interviewing, but you're free to use it for at any step of your recruiting process. There is also more to evaluate in a potential hire than technical skills (for example, cultural match).

Why did I spend 4 years in engineering school, and build my professional track record, yet have to keep doing these negging tests? Does this company not employ anyone who can get a sense of a fellow engineer's skill and professionalism just by talking with them? Did they read my resume? And now they're not even administering the test themselves?

There's two parts to this: (1) On one hand, the interview should never feel like a negging test and the interview of a graduate should be completely different from the interview of a senior engineer with 5+ years of experience. And if you are extremely senior with an amazing professional track record, then perhaps no technical interviewing is necessary at all. (2) On the other hand, no one should feel offended for having to do a technical interview that matches their seniority and experience. It's outsourced to a third party company, because we can do it better and faster. Candidates will be happier, because they can interview outside business hours (incl. weekends) and have a positive experience. Also, the alternative in many cases is that the company won't bother interviewing the candidate at all, or worse, they might send a take-home test that takes hours to 100 candidates and then spend 0-3 minutes reviewing each.

Currently a job-seeker can not get their report card sent to multiple employers, however this is a good idea to consider after getting to a certain scale. At that point, we would essentially become the standard in hardware technical interviewing. Figuring out a fair payment model would be one of the challenges.

Thank you - we think we’re at least 10x better than a non-technical recruiter (also because they have completely different incentives).

These reservations are fair. Hopefully, eventually we’ll have enough endorsements and references to give you some confidence. If you’d like to give us a try before then, happy to talk whenever at founders@touvlo.co!

We don’t do cultural fit at all, and we don’t intend to. I agree this is very important, yet very difficult and inefficient to outsource. We don’t aim to replace all interviewing: becoming a better substitute for early-stage technical screening is more than enough.

Thank you, this is good feedback. We’re both native greek speakers, and we meant it in a lighthearted way, but you’re right that not everyone will see it this way.

You’re absolutely our target market! You’re not delegating the decision, you’re delegating the interviewing. We don’t make a decision, we give you a video and scorecard of a one hour technical interview. You can integrate our interviews at any stage in your pipeline (e.g instead of a phone screen), and do more interviewing for those that do well. Please note that this is common in software, see karat.com.

Yes, we can use any CAD software (or at least, we will seriously consider it). We’re a new company and following the YC advice of doing things that don’t scale. If that means buying 5 different software licenses for our first 5 customers, then we’ll do that.

Thank you!

That’s a good point regarding hardware in the name. Eventually we’d like to do all hardware and more (e.g. architecture), but currently our expertise is in mechanical and robotics engineering.

Our pitch: We make recruiting easier, better and cheaper for the companies building the hardware of tomorrow, by letting them focus on their product rather than repetitive candidate skill assessments. Currently, senior engineers in small startups and scale ups spend a lot of their time interviewing, and they often don’t do it well (remote CAD sessions aren’t really a thing). We save them time and money.

We will scale this by hiring freelancer interview engineers. This has already been done in software, very successfully: karat.com. So we think it can work at least equally well in hardware.

We’re both greek, it’s meant in a joking way :) What do you think of the name?

This is onshape indeed, but we run a collaborative VNC (soon-to-be RDP) session through the browser. We can run anything that runs on Windows (so everything).

Hi! We’re not simulating anything, we run a collaborative VNC (soon-to-be RDP) session through the browser. We can run anything that runs on Windows.

I've spent a lot of time flying on VATSIM/IVAO, and some time controlling. I also eventually got my real-world Private Pilot License.

VATSIM is an amazing community and the quality of the ATC service offered is incredibly high. It definitely helped a lot with becoming more comfortable with using the radio as a real world student pilot. I also expect that it will help me with my instrument license once I decide to do it.

Noting that the project and all its documents are publicly accessible to anyone at https://fbarchive.org/.

A taxiway that goes under a runway would require large, long inclines and declines. An A320 is ~12m tall, and with a 5% incline/decline (which is very large) you'd require 480m of horizontal distance in incline/decline, which sounds like a lot. And some airplanes are two times as tall as the A320, and with some margins added, it's probably safe to guess that an underground passing would require airplanes to taxi hundreds of additional meters to pass through the tunnel.

Airplanes burn a lot of fuel when taxiing, and taxi very slowly. So it's likely an expense, time and space constraint.

The "For once you have tasted flight" quote is well-known tob be mis-attributed to da Vinci:

https://quoteinvestigator.com/2019/01/07/flight/

They gave some justifications for my complaints and apologized. They let me cancel my Metal plan without deducting the cancellation fee, as per my request. They sent me a copy of all my data as I later requested, and deleted my account.

I'm not the OP, but if you're interested in criticisms of Revolut, here's a complaint/rant letter (yes, physical letter) that I sent to them half a year ago: https://pastebin.com/L507bvf7

Revolut has terrible customer support, even for paying customers of the highest tier (and they're super expensive!).

Their in-app chat support agents take ages to respond and are either robots, or they act like they're robots. They know very few things, and will never let you speak to someone who can actually help you. They intentionally hide their email addresses, their "chat agents" always refuse to give you an email and you won't find an email address on their website either. I also tried support@revolut.com and also info@, hello@ and others.

I once had to write a PHYSICAL LETTER that I sent via post to complain. They immediately responded via email (and complied with my requests) from the following address. Feel free to use it to complain:

formalcomplaints@revolut.com

I am honestly curious as to what you found confusing. I think it's fairly well structured and written.

How on Earth is it related to populism?

A populist would generally be opposed to the concentration of wealth and power in a small subset of the population. Populism is on the rise today, as is the trend of family investment offices which are a result of concentrated wealth and power. Thus family offices are destined to face uncomfortable questions by populists.

they didn't create inequality. Yes, you're agreeing with the author. The surrounding sentances from the sentence you quoted:

"[...] the objections to them [family offices] will rise exponentially. The most obvious of these is the least convincing — that family offices have created inequality. They are a consequence, not its cause."

The article does not say that family offices have created inequality, it actually says that they have not.

Regarding the "point of the article":

This is an Economist "Leader", which is a one-page summary of the three-page full article published in the print version of the newspaper. It says the following:

1) Explains what "family offices" are: rich individual's DIY private investment firms.

2) Explains why the populist argument "family offices create inequality" is wrong.

3) Presents three arguments for "family offices are bad" with which the author does not agree, and their rebuttals:

a) "family offices destabilize the financial system" - but data shows that actually they are doing the opposite.

b) "family offices magnify the power of the wealthy" - but this is against the interests of their owners who want to diversify.

c) "family offices might beat regular investors because they have privileged access to information" - they don't outperform the markets right now, but privileged access to information and insider trading could be a problem with family offices.

The author arrives at the conclusion that family offices are a force for good, but more regulation and transparency might be required to address (3c).

This is neither what the article is saying or implying, nor the general view or bias of this newspaper.

With 85% accuracy, 85 of liars are flagged as lying (correctly), [...]

This is not what accuracy means.

85% accuracy just means that 85% of all the decisions the system makes are correct. A system in such a setting, where a single false negative matters a lot more than a single false positive (which would simply be handed over to a human for further investigation) would necessarily be tuned for extremely high recall at the cost of precision. In other words, it would often flag innocent people for further investigation (as you've said), but it would almost never clear people that should've been flagged.

It hasn't just slowed down. MAU/DAUs in Europe have dropped compared to last quarter, and stayed flat in the US.

11% may be the YoY growth in world-wide MAU, but MAU has actually dropped in Europe and remained constant in the US compared to last quarter.

As far as I know, her videos were demonetized, not rejected. There's a huge difference between being silenced and being refused ad placement on your videos.

Getting paid to express your opinion is not a human right.

Google pays 120k+ for junior developers in Switzerland.

Very relevant paper: "Hardened dildo.io, A Cryptographically Secure, Usable Matchmaking Service"

https://courses.csail.mit.edu/6.857/2016/files/13.pdf

I've actually implemented this paper in a hackathon, but we never launched it. It's basically like tinder, but no central server has access to anyone's likes. Yet, by using homomorphic encryption and a calculation performed on the server (through which no information is revealed), you can know if someone you liked likes you back. It's pretty cool!

In the US you have the opportunity to reach for the stars (literally). If you want it bad enough, with a little luck and hard work you can get there.

Not true. The US ranks very poorly compared to other advanced economies when it comes to income inequality and social mobility [1].

Also most of the examples you give prove exactly this: Mark Zuckerberg was just a kid in a Harvard dorm, and so was Bill Gates (one of your "examples"). Ted Turner went to Brown, and all almost every person in the list you gave went to a very good school, if not Ivy League. And yet most Americans claim that they can't afford college [2]. It is naive to claim that "most people aren't rich due to a lack of opportunity", because that's exactly what's happening.

[1]: "How to get rich in America", The Economist -http://www.economist.com/blogs/economist-explains/2017/02/ec...

[2]: "83% of Americans say they can't afford college", Edward Jones study -http://www.thinkadvisor.com/2015/05/13/83-of-americans-say-t...

I just tried it, seems to work fine.