This article is excessively LLM written to the point that there’s barely any real information in between all of the unnecessary pie charts and repeated points.
This move is being executed too broadly, in my opinion, but the “bad labs” problem especially in China is widely known in the industry. If you spend any time in the electronics industry at smaller companies you will encounter people who know Chinese labs that will give your product passing test results every time as long as you’re not so far past the limits that it’s too obvious.
Yes when qualifying new equipment often the source of failing emissions is the power supply (or some other third party device) that has already 'passed' testing.
The search site they point to (https://markready.io/labs) also lists incorrect locations on the map for labs. BUREAU VERITAS CONSUMER PRODUCTS SERVICES, INC. shows on the map in Littleton, Colorado but is actually located in Littleton, Massachusetts.
i think it's better to flag the article and move on. the author's replies in the thread are also LLM authored. and nothing in the article makes any sense.
LLMs often have a distinct writing style. It's not guaranteed, you can get false positives and false negatives, but if you start paying attention it becomes obvious in many cases.
That obviously won't be true for much longer, assuming it's still true now, which I doubt. If you're an LLM content farmer, how hard could it possibly be to LoRA your way out of generating cliches like emdashes, 'You're absolutely right!' and 'It's not A, but B' rhetoric?
But you can tell it to use different styles. To be formal or in-formal, to insert colloquialisms or to remove.
People are depending on their own 'gut-sense' a lot, and not realizing they are really not correct.
If you think all it takes is paying attention, then you are missing it. It's both more widely used than assumed, and also now obscuring what is non-AI.
> But you can tell it to use different styles. To be formal or in-formal, to insert colloquialisms or to remove.
And when you get it right, the result doesn't get called AI generated.
> People are depending on their own 'gut-sense' a lot, and not realizing they are really not correct.
TFA is very obvious about it.
A human who writes like this should be ashamed to do so, and should endeavour to understand why the writing comes across as "generic LLM"-like and fix it.
We have reached a point where people can end up training their writing on generic LLM output. This is a bad thing, because it's bad output.
Even beyond any clues from writing style, the general presentation is bad. It presents far too many facts and figures without giving anyone a good reason to care about most of them. And then it ends with a section on a separate topic (how to choose a lab, rather than how they're distributed across the world).
Most importantly, though, the submission is presented with a different title that implies a different purpose to the article that is not elaborated in the article. I would have expected personal insight a) on why people should care about the FCC's action (there is no mention of that action at all); b) on what the process was like of collecting this data. And I would have expected, you know, mapping of the lab locations rather than bar charts giving geographic breakdowns.
Don't beat yourself over it. It's the new sport for HN upvote farmers to default to calling out any TLDR post that got "delve" in it or some other cliché as LLM Slop. I also think it's a waste of time. What's important is the content. Is the content of the article valuable? No? Just close it and move on. But we know the incentives to get a few upvotes is just much too good to pass...
This article goes ham on the rule of threes, it does the "not just x, but y" cliche, em-dash with spaces on either side, bold heading-sentence paragraphs, it visibly has hallmarks of AI driven writing.
If you personally can't tell then just say that rather than casting aspersions on everyone else by claiming they can't.
No human* would waste the time to write a piece that is both highly polished while being so long that any useful information is spread so thinly it is essentially empty. This is how people "can tell" if it is written by AI.
Not a dig at this author by the way or saying it applies to this post, just in general.
*or if they did anyway, the result is the same: bad writing.
> a piece that is both long and highly polished while being devoid of useful information
Idk, I learned a little bit about our regulatory system, that a lot of these labs are in China and that those are now banned (and that the ones in India may be next).
The style is admittedly annoying. But I'm glad the author put in the work to highlight something they, and now I through them, found interesting.
No human would waste the time to write a piece that is both highly polished while being so long that any useful information is spread so thinly it is essentially empty.
LOL, some of us spent 12 years in public schools refining this very art to perfection.
> thought they were visually helpful in this instance
If you're the author, can you comment on whether you used AI to write this? (Specifically, the text.)
Where it might be suffering is in its presentation of a list of facts unorganised around any thesis. It took me until your China Question section to see the meat of your piece.
If I had to suggest some edits, they would be making everything above that section more concise (by reducing the number of charts and/or moving them to footnotes) and adding a summarising subtitle.
There are also jargon jumps, e.g. from TFAB to TCB. (I initially assumed the FAA was a TCB, the latter being a generic international term.) This compounds the lack of conciseness presented by the accredition-body breakdown and TCBs vc. test-only labs sections. If those sections were moved after your thesis section, you could dive into whether China's labs differ from the U.S. labs in those respects.
The content of the site is, as stated in my first comment and in the article itself, a nice looking wrapper on top of in essence, an llm Wiki that I put together with the help of Claude on the hardware certification universe. While I was building this data set out, I uncovered that the FCC had this vote today, so I thought it would be a good thing to share since it's timely and because I had just collected all of the relevant information tolp someone figure out how this impacts their hardware certification process (I use voice transcription to write this comment)
I very much appreciate your feedback. As I look at the article now. I totally see what you're saying. I should have let off what was going on with the vote today since that's what I referenced in the title of the post on here.
the headline hints that there's some sort of non-obvious factor that's going to be revealled. I scrolled past countless redundant and information-sparse graph-like figures and never found it; moved on.
If you've only got a paragraph worth of information to share, say it and let us get on with our lives.
Well I couldn't find any other thorough dataset on this topic, so in that sense this is non-obvious since it took weeks to assemble the information. And it was fun doing it using the LLM Wiki technique.
The general idea is to have the LLM maintain longer-term context/background by storing it in a format/structure that's akin to a standard Wiki. The result is (hopefully) a series of human-readable and editable documents that's developed and maintained by the agent.
Git + claude code in yolo mode. In the first prompt, I passed it Kaparthy's gist, and had it put together a high level plan of all of the sections that needed to be written to complete a vision I provided. Essentially put together a complete wiki on everything for getting global hardware certification.
I then had it loop once an hour. It would pick the next wiki to write, research it, gather raw sources, and then synthesize the wiki for me and push. I could nudge it in between hours if I wanted.
Do you add this as a skill or permanent prompt, making it always maintain wikis in the background? Or do you direct it to make these only when you're in the project?
There's skills if you want. I didn't want to do that since I don't feel like I need the smarts to work on the LLM wiki in every coding session. I like to keep my context clean and scoped to what I'm working on.
I am running this in a long running session that spawns a subagent once an hour. So the context of the main session doesn't get out of control.
> The core proposal bans all labs in China and Hong Kong, extending the "Bad Labs" rule the FCC adopted in May 2025. A broader proposal would also cut off labs in any country without a Mutual Recognition Agreement with the US, which adds 5 more labs (4 in India, 1 in Switzerland). Total at risk: 131 labs, 22.2% of the global total.
As expected this is an attack on China-sourced electronics.
I hope this isn’t the start of a return to the Bad Times when many niche electronics were simply unavailable at any price, and what was available was $1K+ for what should be commodity gear.
What the FCC does is important, but there needs to be a sense of proportionality. I am a ham radio user but I am not particularly bothered if my $30 DVD player has a few spurious emissions, as long as they aren’t egregious. I also don’t mind imperfect but cheap radios like Baofengs if they help get people into the hobby. It’s good to have a box of these to hand out in emergency situations! Can’t do that with Yaesus unless you’re made of money.
> What the FCC does is important, but there needs to be a sense of proportionality. I am a ham radio user but I am not particularly bothered if my $30 DVD player has a few spurious emissions, as long as they aren’t egregious. I also don’t mind imperfect but cheap radios like Baofengs if they help get people into the hobby. It’s good to have a box of these to hand out in emergency situations! Can’t do that with Yaesus unless you’re made of money.
I'm bothered when my neighbors turn on their christmas lights, and the whole 40 meter band is wiped out.
Also baofengs are horrible all those regards:
* spurious emissions (thus banned in quite a few countries)
* useless in most emergencies (but preppers somehow buy them for some reason... probably due to youtubers shilling for them)
* handing them out to whom exactly? You need a ham radio licence to use them, and i'm pretty sure every licenced ham has a radio and doesn't need handouts from others (unless we're talking about baofeng FRS/PMR radios, but somehow preppers never buy those)
Also a yaesu ft65 costs around 100eur over here, you don't have to be made of money to afford a much better radio.
You don’t actually need a license to use the ham bands in a true emergency.
Where I’m at a Baofeng can hit the local repeaters just fine. I handed them out to my family when we had a major multiday communication outage (cellular and internet were down) and set them up to listen to the repeater. I told them if there’s a life threatening emergency they can transmit. It made everyone feel a little safer.
While I personally have a better radio, they are great as cheap backups.
Legally you do, that exception only applies to amateur station, not unlicenced users.
Why not get a gmrs licence instead, and give them gmrs type-accepted radios that they can use and try out and get experienced with even when not in an emergency? It's like buying cheap cars to give to people to drive for the first time in an active emergency... dangerous both to them and to others.
I’m afraid you are misinformed, FCC Part 97.403 allows unrestricted use in a true life threatening emergency. The words “amateur station” may be misleading here, they apply to the equipment not the licensed individual. See this discussion for more information: https://old.reddit.com/r/amateurradio/comments/1fyhp9f/lets_...
GMRS may not be an option if there are no repeaters in your area.
No, the equipment is the "apparatus". Only when used by a licenced ("duly authorized") individual for the use of (...), that equipment becomes an amateur station.
For example, a part 90 type-accepted business radio is just a business radio, part 97 has no ruling over it nor the user nor nothing related to that radio. Once a licenced amateur uses it for amateur radio stuff (so that radio is in amateur service), it becomes an amateur station and governed by the rules in part 95.
I'd wager Baofeng is the most common emergency radio. Baofeng or something equivalent is what people in the 3rd world have largely been able to actually afford and there in the rough that's actually what's being used. I recall Baofengish radios being the most commonly spotted ones in the Syrian Civil War.
But it's not an emergency radio, it's a cheap chinese radio that's... well. cheap. That's like buying a $60 android smartphone from aliexpress... and yes, many people in 3rd world countries buy those too.
The frontend is horrible, the filtering is horrible, they get easily overloaded, and they're still not emergency radios. People will die because they will rely on them instead of getting a proper radio for emergencies. Garmin inreach will actually get you help when stuck in the middle of nowhere (because no one will be in simplex range then), and a starlink setup is much better for anything at home, becuase you can actually reach someone who can help that way. Baofengs are just something that earns percentages to youtube "preppers" (many of them not licenced hams either... it's like taking car advice from someone who doesn't even have a drivers licence).
Maybe someone here will start manufacturing these things?
Of course initially, expecting the same quality and low price will likely be an issue. But over time it gets better is probably the idea. Will it actually happen, who can say? But I can understand the idea they have here. I'm not saying they'll be successful. Definitely not saying I agree with it. (There are far more effective ways to accomplish a manufacturing ramp up with far less risk.) But I get the idea.
The economics just don't really make sense it's so much cheaper to produce abroad and ship it here, also this isn't manufacturing just the testing labs and these exist to provide rubber stamps of good enough products that maybe have a few issues.
I would love to see reshoring, but we have to be realistic about it and target the areas most critical to sovereignty and security. The scale of Chinese manufacturing is mind boggling, and we probably aren’t going to be making random affordable consumer products any time soon.
Maybe the FCC lab rules ought to be more selective, allowing consumer goods but not commercial/industrial goods at these labs.
Alternatively, maybe we can subsidize the consumer. What I don’t want to see is everything becoming 10x more expensive or completely unavailable at home while the rest of the world gets to keep the status quo.
Yep the only real area that reshoring can provide savings is in shipping and iteration time. The former is cheap compared to US labor and the latter can be solved by having your design team in China too.
I have first hand experience with this. It may look like a politically motivated move, but in the RF industry when you need something to pass, China is known to be much friendlier/lax than other labs.
For those who don't know, the point of these labs is to generate certified test results. The lab's job is to certify that the test was done correctly with calibrated test equipment. They then give you the results which you then give to the regulating body.
However, you are the one who knows your product and how it operates. You are present in the lab, in the room, doing the testing with them. This introduces a lot of grey area where a lab may or may not go along with what you say. Chinese labs are known for just going along with it. After all, you are the one paying them tens of thousands, and they know you probably have many other products that need certifying. It's mutually beneficial for them to be lax.
> Chinese labs are known for just going along with it. After all, you are the one paying them tens of thousands,
Are the engineering firms going to Chinese labs because of their rubber stamp approval or because the process is “tens of thousands” as opposed to “hundreds of thousands”?
It's cheaper and they are friendlier. They don't rubber stamp, the tests are still done, but if you are struggling to get over a .5dB hurdle they will "work with you" to snuff it out.
The tests can be 100+ parameters and every single one is a hard binary pass/fail. So a mouse fart over the line on one test and you are not compliant. Go home, re-engineer, hope you fixed the issue (it's basically impossible to test for without a full dedicated RF testing chamber), and pay for a whole new testing run.
This is where the Chinese labs come in to play, as upon seeing your .5dB issue, they may feel that maybe their antenna placement was slightly off, or maybe you say that actually the "center of noise" (a bs thing) on your product is 5cm away from the testing center point, and move it back from the antenna with a wink and a nod.
In western labs the people are just as friendly, but that antenna is definitely in the right spot, and your product goes in the same spot as everyone elses.
However you need to go to China to do all this. If you are big megacorp with huge resources, western labs are fine. If you are smaller and possibly will become insolvent if you cannot pass testing, China is the play.
As far as I can tell, this distinction is mostly meaningless in practice.
Most people are going to want to sell their products on an international market, which essentially means designing and testing it to the strictest rules any country uses and just having to do a whole bunch of paperwork for the rest: getting a lab to do both FCC and CE is not significantly more expensive than either one on its own. And because FCC requires external testing, that means doing external testing.
Besides, although CE is technically a self-declaration that you follow the relevant rules, it still requires you to be able to demonstrate that you follow those rules - which means you have to test and report on a level comparable to an external lab, which means building a testing lab with a price tag comparable to a very nice home, and doing all the annoying paperwork like having your equipment regularly tested and calibrated. You are allowed to do it in-house, but is it worth it?
Yeah, this article is really focused on FCC certification. But to your point, there are other sections on the site that are focused on CE certification and how to navigate getting certified so you can sell throughout the world.
Yes thats evidently clear but you usually don’t need external accreditation if you do in house testing and keep a compliance file for the product, this applies for both the US and EU.
In-house emc testing is quite fun and you dont need much more than a spectrum analyzer, antenna and E/H-field probes.
In-house EMC testing is practically always required before you go to an external lab, unless you want to waste a lot of money by discovering at the lab that your device cannot be certified.
When done just for this purpose, it can be done much more cheaply than at a proper lab, because you do not need very accurate results.
Well there's a bit more to it than that. It depends on if you're making an intentional radiator. I have another flow chart on the site that helps you figure out if you need to send your device to a testing lab or not.
This will make testing more expensive and hence some will no longer bother to FCC test excluding the US from receiving their product. Especially small companies and startups.
I did not FCC certify my product because the primary focus is for the European market but now I would also have to consider costs.
Probably because it appears to have been written by an LLM (the post too, but I assume comments are more easily killed by flags (?)). (To be clear I did not flag it myself.)
I've been building a certification intelligence tool for hardware teams (markready.io) and needed a good test lab directory. The FCC publishes accreditation data through a Socrata API but it's pretty bare - names, addresses, designation numbers, and expiration dates that are often years stale. No websites, no capabilities, nothing to tell you whether a lab is a two-person shop or an Intertek subsidiary.
The first thing I did was build an LLM-maintained wiki about the hardware certification universe - FCC rule parts, equipment authorization types, test standards, the TCB system, international equivalents, all of it. About 30 pages of structured knowledge that Claude could reference when doing the actual enrichment work. Then I ran a loop of subagents over multiple days to enrich the labs - pulling from the Socrata API, cross-referencing TCB registrations to see which labs can certify (not just test), hitting Google Places for websites and coordinates, crawling accreditation body directories to figure out which labs are actually still active. The FCC's own expiration dates are useless for this - tons of labs show 2022 or 2023 dates but are clearly still operating. Claude synthesized the descriptions and capabilities from all the scraped data into structured records, using the wiki as context.
The directory is at markready.io/labs. You can browse by country, US state, and TCB status.
Today's an interesting day to launch this because the FCC is voting to ban all 126 test labs in China and Hong Kong. Not just the government-controlled ones they banned last year - all of them. 21% of the global total gone. 27 of those 126 are Western firms (Intertek, SGS, TUV, UL) operating China offices. I wrote up the full impact analysis at markready.io/blog/fcc-bad-labs-vote.
Full disclosure: I've never actually gotten a device through FCC certification myself. I've been building RF hardware since I was a kid but always on the hobby side. What pulled me into this was the data problem - the FCC publishes all this information but nobody had stitched it together into anything usable, and it felt like a genuinely interesting dataset to enrich and a real gap in understanding the hardware product space.
Built with Next.js + Cloudflare. The enriched dataset covers 28 countries.
> Today's an interesting day to launch this because the FCC is voting to ban all 126 test labs in China and Hong Kong
Does this increase or decrease demand for your tool? Less fragmentation would be expected to decrease demand from insiders. But more regulatory scrutiny would raise the stakes for outsiders getting it wrong.
I think this increases the relevancy for these tools and information. Gone will be the days of just sending your design to manufacturer in China and having it get fully certified and built through just one contact.
At this point I wish it were against the rules to accuse people or complain about articles as written by LLMs. It's creating so much noise that useful commentary is hard to find.
I don't see any signs of the parent comment being written by an LLM other than it's detailed and well-written.
Their comment does not give off "LLM-written" really... It drives forwards actual points without superfluous segments. I don't think it's helpful to try and discredit people whenever we want by throwing around accusations of "LLM-written".
This article is excessively LLM written to the point that there’s barely any real information in between all of the unnecessary pie charts and repeated points.
This move is being executed too broadly, in my opinion, but the “bad labs” problem especially in China is widely known in the industry. If you spend any time in the electronics industry at smaller companies you will encounter people who know Chinese labs that will give your product passing test results every time as long as you’re not so far past the limits that it’s too obvious.
Yes when qualifying new equipment often the source of failing emissions is the power supply (or some other third party device) that has already 'passed' testing.
Interesting so like a power supply module that has already gotten full modular approval? Do you have an example link to one of those? Thanks
The search site they point to (https://markready.io/labs) also lists incorrect locations on the map for labs. BUREAU VERITAS CONSUMER PRODUCTS SERVICES, INC. shows on the map in Littleton, Colorado but is actually located in Littleton, Massachusetts.
Thanks for finding a data quality issue and reporting it.
Is that why the pie chart colors don't match the colors in the legend at all..?
Thanks for catching that, fixed! Legend didnt match because I made a mistake when I tried to clean up the contrast in the pie chart.
i think it's better to flag the article and move on. the author's replies in the thread are also LLM authored. and nothing in the article makes any sense.
[flagged]
I've noticed a trend of calling every single article "This is AI or LLM, I can't stand it".
And really, you can't tell. Nobody can tell. Humans write badly and blandly also. It's just a trope at this point.
No, you're comment is an LLM.
LLMs often have a distinct writing style. It's not guaranteed, you can get false positives and false negatives, but if you start paying attention it becomes obvious in many cases.
My poor Reddit has been taken over by bots :(
Reddit is extra cooked soon.
I’m going to assume everyone using “cooked” comes from uThermal and you won’t convince me otherwise.
That obviously won't be true for much longer, assuming it's still true now, which I doubt. If you're an LLM content farmer, how hard could it possibly be to LoRA your way out of generating cliches like emdashes, 'You're absolutely right!' and 'It's not A, but B' rhetoric?
We should probably go ahead and get over it.
Yes, if you are using a generic LLM.
But you can tell it to use different styles. To be formal or in-formal, to insert colloquialisms or to remove.
People are depending on their own 'gut-sense' a lot, and not realizing they are really not correct.
If you think all it takes is paying attention, then you are missing it. It's both more widely used than assumed, and also now obscuring what is non-AI.
> But you can tell it to use different styles. To be formal or in-formal, to insert colloquialisms or to remove.
And when you get it right, the result doesn't get called AI generated.
> People are depending on their own 'gut-sense' a lot, and not realizing they are really not correct.
TFA is very obvious about it.
A human who writes like this should be ashamed to do so, and should endeavour to understand why the writing comes across as "generic LLM"-like and fix it.
We have reached a point where people can end up training their writing on generic LLM output. This is a bad thing, because it's bad output.
Even beyond any clues from writing style, the general presentation is bad. It presents far too many facts and figures without giving anyone a good reason to care about most of them. And then it ends with a section on a separate topic (how to choose a lab, rather than how they're distributed across the world).
Most importantly, though, the submission is presented with a different title that implies a different purpose to the article that is not elaborated in the article. I would have expected personal insight a) on why people should care about the FCC's action (there is no mention of that action at all); b) on what the process was like of collecting this data. And I would have expected, you know, mapping of the lab locations rather than bar charts giving geographic breakdowns.
I guess my point was lost.
It is obvious, when it is obvious. When it is not, you don't know it.
There are ton more false positives now. Everyone is calling everything 'LLM Slop'.
Because there is a lot of slop. Now every bad human writer is being called an AI just for being human.
And, that is covering that a ton of stuff is LLM and nobody can tell.
People that say they can tell the difference are fooling themselves.
Don't beat yourself over it. It's the new sport for HN upvote farmers to default to calling out any TLDR post that got "delve" in it or some other cliché as LLM Slop. I also think it's a waste of time. What's important is the content. Is the content of the article valuable? No? Just close it and move on. But we know the incentives to get a few upvotes is just much too good to pass...
Thank You. I'm not the only one that saw the new pattern.
This article goes ham on the rule of threes, it does the "not just x, but y" cliche, em-dash with spaces on either side, bold heading-sentence paragraphs, it visibly has hallmarks of AI driven writing.
If you personally can't tell then just say that rather than casting aspersions on everyone else by claiming they can't.
Fun fact, the author admits to using a LLM.
https://news.ycombinator.com/item?id=47963465
Not the first time this has happened.
Half the articles are now LLM's.
If he didn't admit it, we'd be arguing over 'style', which itself can be configured.
Prompt> LLM don't use em-dash.
LLM> OK.
No human* would waste the time to write a piece that is both highly polished while being so long that any useful information is spread so thinly it is essentially empty. This is how people "can tell" if it is written by AI.
Not a dig at this author by the way or saying it applies to this post, just in general.
*or if they did anyway, the result is the same: bad writing.
> a piece that is both long and highly polished while being devoid of useful information
Idk, I learned a little bit about our regulatory system, that a lot of these labs are in China and that those are now banned (and that the ones in India may be next).
The style is admittedly annoying. But I'm glad the author put in the work to highlight something they, and now I through them, found interesting.
No human would waste the time to write a piece that is both highly polished while being so long that any useful information is spread so thinly it is essentially empty.
LOL, some of us spent 12 years in public schools refining this very art to perfection.
Haha, good one Altman!
I wake up, there is another psyop, I go to sleep
Thanks for that insight on how the Chinese labs are perceived amongst hardware engineers.
I know pie charts are decisive. I thought they were visually helpful in this instance.
> thought they were visually helpful in this instance
If you're the author, can you comment on whether you used AI to write this? (Specifically, the text.)
Where it might be suffering is in its presentation of a list of facts unorganised around any thesis. It took me until your China Question section to see the meat of your piece.
If I had to suggest some edits, they would be making everything above that section more concise (by reducing the number of charts and/or moving them to footnotes) and adding a summarising subtitle.
There are also jargon jumps, e.g. from TFAB to TCB. (I initially assumed the FAA was a TCB, the latter being a generic international term.) This compounds the lack of conciseness presented by the accredition-body breakdown and TCBs vc. test-only labs sections. If those sections were moved after your thesis section, you could dive into whether China's labs differ from the U.S. labs in those respects.
The content of the site is, as stated in my first comment and in the article itself, a nice looking wrapper on top of in essence, an llm Wiki that I put together with the help of Claude on the hardware certification universe. While I was building this data set out, I uncovered that the FCC had this vote today, so I thought it would be a good thing to share since it's timely and because I had just collected all of the relevant information tolp someone figure out how this impacts their hardware certification process (I use voice transcription to write this comment)
I very much appreciate your feedback. As I look at the article now. I totally see what you're saying. I should have let off what was going on with the vote today since that's what I referenced in the title of the post on here.
the headline hints that there's some sort of non-obvious factor that's going to be revealled. I scrolled past countless redundant and information-sparse graph-like figures and never found it; moved on.
If you've only got a paragraph worth of information to share, say it and let us get on with our lives.
Well I couldn't find any other thorough dataset on this topic, so in that sense this is non-obvious since it took weeks to assemble the information. And it was fun doing it using the LLM Wiki technique.
> it was fun doing it using the LLM Wiki technique
What is that?
The general idea is to have the LLM maintain longer-term context/background by storing it in a format/structure that's akin to a standard Wiki. The result is (hopefully) a series of human-readable and editable documents that's developed and maintained by the agent.
There's great coverage of it at https://gist.github.com/karpathy/442a6bf555914893e9891c11519...
It's actually also now a base capability in the Hermes agent and has been really helpful for me, at least.
https://gist.github.com/karpathy/442a6bf555914893e9891c11519...
Could you share the tools you used to do this? I'd love to organise my esoteric research side quests like this.
Git + claude code in yolo mode. In the first prompt, I passed it Kaparthy's gist, and had it put together a high level plan of all of the sections that needed to be written to complete a vision I provided. Essentially put together a complete wiki on everything for getting global hardware certification.
I then had it loop once an hour. It would pick the next wiki to write, research it, gather raw sources, and then synthesize the wiki for me and push. I could nudge it in between hours if I wanted.
Do you add this as a skill or permanent prompt, making it always maintain wikis in the background? Or do you direct it to make these only when you're in the project?
There's skills if you want. I didn't want to do that since I don't feel like I need the smarts to work on the LLM wiki in every coding session. I like to keep my context clean and scoped to what I'm working on.
I am running this in a long running session that spawns a subagent once an hour. So the context of the main session doesn't get out of control.
> The core proposal bans all labs in China and Hong Kong, extending the "Bad Labs" rule the FCC adopted in May 2025. A broader proposal would also cut off labs in any country without a Mutual Recognition Agreement with the US, which adds 5 more labs (4 in India, 1 in Switzerland). Total at risk: 131 labs, 22.2% of the global total.
As expected this is an attack on China-sourced electronics.
I hope this isn’t the start of a return to the Bad Times when many niche electronics were simply unavailable at any price, and what was available was $1K+ for what should be commodity gear.
What the FCC does is important, but there needs to be a sense of proportionality. I am a ham radio user but I am not particularly bothered if my $30 DVD player has a few spurious emissions, as long as they aren’t egregious. I also don’t mind imperfect but cheap radios like Baofengs if they help get people into the hobby. It’s good to have a box of these to hand out in emergency situations! Can’t do that with Yaesus unless you’re made of money.
We still as amateur radio operators have the "kit" exemption that comes with 3 components we plug together right?
> What the FCC does is important, but there needs to be a sense of proportionality. I am a ham radio user but I am not particularly bothered if my $30 DVD player has a few spurious emissions, as long as they aren’t egregious. I also don’t mind imperfect but cheap radios like Baofengs if they help get people into the hobby. It’s good to have a box of these to hand out in emergency situations! Can’t do that with Yaesus unless you’re made of money.
I'm bothered when my neighbors turn on their christmas lights, and the whole 40 meter band is wiped out.
Also baofengs are horrible all those regards:
* spurious emissions (thus banned in quite a few countries) * useless in most emergencies (but preppers somehow buy them for some reason... probably due to youtubers shilling for them) * handing them out to whom exactly? You need a ham radio licence to use them, and i'm pretty sure every licenced ham has a radio and doesn't need handouts from others (unless we're talking about baofeng FRS/PMR radios, but somehow preppers never buy those)
Also a yaesu ft65 costs around 100eur over here, you don't have to be made of money to afford a much better radio.
You don’t actually need a license to use the ham bands in a true emergency.
Where I’m at a Baofeng can hit the local repeaters just fine. I handed them out to my family when we had a major multiday communication outage (cellular and internet were down) and set them up to listen to the repeater. I told them if there’s a life threatening emergency they can transmit. It made everyone feel a little safer.
While I personally have a better radio, they are great as cheap backups.
(assuming you're american)
Legally you do, that exception only applies to amateur station, not unlicenced users.
Why not get a gmrs licence instead, and give them gmrs type-accepted radios that they can use and try out and get experienced with even when not in an emergency? It's like buying cheap cars to give to people to drive for the first time in an active emergency... dangerous both to them and to others.
I’m afraid you are misinformed, FCC Part 97.403 allows unrestricted use in a true life threatening emergency. The words “amateur station” may be misleading here, they apply to the equipment not the licensed individual. See this discussion for more information: https://old.reddit.com/r/amateurradio/comments/1fyhp9f/lets_...
GMRS may not be an option if there are no repeaters in your area.
No, the equipment is the "apparatus". Only when used by a licenced ("duly authorized") individual for the use of (...), that equipment becomes an amateur station.
For example, a part 90 type-accepted business radio is just a business radio, part 97 has no ruling over it nor the user nor nothing related to that radio. Once a licenced amateur uses it for amateur radio stuff (so that radio is in amateur service), it becomes an amateur station and governed by the rules in part 95.
I'd wager Baofeng is the most common emergency radio. Baofeng or something equivalent is what people in the 3rd world have largely been able to actually afford and there in the rough that's actually what's being used. I recall Baofengish radios being the most commonly spotted ones in the Syrian Civil War.
Early on they were spotted in the Ukranian conflict as well, before everyone got proper radios.
But it's not an emergency radio, it's a cheap chinese radio that's... well. cheap. That's like buying a $60 android smartphone from aliexpress... and yes, many people in 3rd world countries buy those too.
The frontend is horrible, the filtering is horrible, they get easily overloaded, and they're still not emergency radios. People will die because they will rely on them instead of getting a proper radio for emergencies. Garmin inreach will actually get you help when stuck in the middle of nowhere (because no one will be in simplex range then), and a starlink setup is much better for anything at home, becuase you can actually reach someone who can help that way. Baofengs are just something that earns percentages to youtube "preppers" (many of them not licenced hams either... it's like taking car advice from someone who doesn't even have a drivers licence).
73 fellow ham radio operator!
Maybe someone here will start manufacturing these things?
Of course initially, expecting the same quality and low price will likely be an issue. But over time it gets better is probably the idea. Will it actually happen, who can say? But I can understand the idea they have here. I'm not saying they'll be successful. Definitely not saying I agree with it. (There are far more effective ways to accomplish a manufacturing ramp up with far less risk.) But I get the idea.
The economics just don't really make sense it's so much cheaper to produce abroad and ship it here, also this isn't manufacturing just the testing labs and these exist to provide rubber stamps of good enough products that maybe have a few issues.
I would love to see reshoring, but we have to be realistic about it and target the areas most critical to sovereignty and security. The scale of Chinese manufacturing is mind boggling, and we probably aren’t going to be making random affordable consumer products any time soon.
Maybe the FCC lab rules ought to be more selective, allowing consumer goods but not commercial/industrial goods at these labs.
Alternatively, maybe we can subsidize the consumer. What I don’t want to see is everything becoming 10x more expensive or completely unavailable at home while the rest of the world gets to keep the status quo.
Yep the only real area that reshoring can provide savings is in shipping and iteration time. The former is cheap compared to US labor and the latter can be solved by having your design team in China too.
I have first hand experience with this. It may look like a politically motivated move, but in the RF industry when you need something to pass, China is known to be much friendlier/lax than other labs.
For those who don't know, the point of these labs is to generate certified test results. The lab's job is to certify that the test was done correctly with calibrated test equipment. They then give you the results which you then give to the regulating body.
However, you are the one who knows your product and how it operates. You are present in the lab, in the room, doing the testing with them. This introduces a lot of grey area where a lab may or may not go along with what you say. Chinese labs are known for just going along with it. After all, you are the one paying them tens of thousands, and they know you probably have many other products that need certifying. It's mutually beneficial for them to be lax.
> Chinese labs are known for just going along with it. After all, you are the one paying them tens of thousands,
Are the engineering firms going to Chinese labs because of their rubber stamp approval or because the process is “tens of thousands” as opposed to “hundreds of thousands”?
It's cheaper and they are friendlier. They don't rubber stamp, the tests are still done, but if you are struggling to get over a .5dB hurdle they will "work with you" to snuff it out.
The tests can be 100+ parameters and every single one is a hard binary pass/fail. So a mouse fart over the line on one test and you are not compliant. Go home, re-engineer, hope you fixed the issue (it's basically impossible to test for without a full dedicated RF testing chamber), and pay for a whole new testing run.
This is where the Chinese labs come in to play, as upon seeing your .5dB issue, they may feel that maybe their antenna placement was slightly off, or maybe you say that actually the "center of noise" (a bs thing) on your product is 5cm away from the testing center point, and move it back from the antenna with a wink and a nod.
In western labs the people are just as friendly, but that antenna is definitely in the right spot, and your product goes in the same spot as everyone elses.
However you need to go to China to do all this. If you are big megacorp with huge resources, western labs are fine. If you are smaller and possibly will become insolvent if you cannot pass testing, China is the play.
Thank you for these amazing insights.
This article completely omits a section on when external accreditation. For many products you can get CE marking or similar with in-house testing.
As far as I can tell, this distinction is mostly meaningless in practice.
Most people are going to want to sell their products on an international market, which essentially means designing and testing it to the strictest rules any country uses and just having to do a whole bunch of paperwork for the rest: getting a lab to do both FCC and CE is not significantly more expensive than either one on its own. And because FCC requires external testing, that means doing external testing.
Besides, although CE is technically a self-declaration that you follow the relevant rules, it still requires you to be able to demonstrate that you follow those rules - which means you have to test and report on a level comparable to an external lab, which means building a testing lab with a price tag comparable to a very nice home, and doing all the annoying paperwork like having your equipment regularly tested and calibrated. You are allowed to do it in-house, but is it worth it?
Yeah, this article is really focused on FCC certification. But to your point, there are other sections on the site that are focused on CE certification and how to navigate getting certified so you can sell throughout the world.
Yes thats evidently clear but you usually don’t need external accreditation if you do in house testing and keep a compliance file for the product, this applies for both the US and EU.
In-house emc testing is quite fun and you dont need much more than a spectrum analyzer, antenna and E/H-field probes.
In-house EMC testing is practically always required before you go to an external lab, unless you want to waste a lot of money by discovering at the lab that your device cannot be certified.
When done just for this purpose, it can be done much more cheaply than at a proper lab, because you do not need very accurate results.
Well there's a bit more to it than that. It depends on if you're making an intentional radiator. I have another flow chart on the site that helps you figure out if you need to send your device to a testing lab or not.
This will make testing more expensive and hence some will no longer bother to FCC test excluding the US from receiving their product. Especially small companies and startups.
I did not FCC certify my product because the primary focus is for the European market but now I would also have to consider costs.
This is a valid concern. For the time being I'd expect Taiwanese labs to be a price comparable alternative, but this is a likely outcome.
Why is OPs comment on his own work that hit first page flagged and dead? Is it repeating info from the article?
Probably because it appears to have been written by an LLM (the post too, but I assume comments are more easily killed by flags (?)). (To be clear I did not flag it myself.)
It's unflagged and undead now, so probably not because of that but a mistake? And the comment is coming for our brains too
Clank clank, beep boop. You should walk to the car wash since its only 30m away ;)
I've been building a certification intelligence tool for hardware teams (markready.io) and needed a good test lab directory. The FCC publishes accreditation data through a Socrata API but it's pretty bare - names, addresses, designation numbers, and expiration dates that are often years stale. No websites, no capabilities, nothing to tell you whether a lab is a two-person shop or an Intertek subsidiary.
The first thing I did was build an LLM-maintained wiki about the hardware certification universe - FCC rule parts, equipment authorization types, test standards, the TCB system, international equivalents, all of it. About 30 pages of structured knowledge that Claude could reference when doing the actual enrichment work. Then I ran a loop of subagents over multiple days to enrich the labs - pulling from the Socrata API, cross-referencing TCB registrations to see which labs can certify (not just test), hitting Google Places for websites and coordinates, crawling accreditation body directories to figure out which labs are actually still active. The FCC's own expiration dates are useless for this - tons of labs show 2022 or 2023 dates but are clearly still operating. Claude synthesized the descriptions and capabilities from all the scraped data into structured records, using the wiki as context.
The directory is at markready.io/labs. You can browse by country, US state, and TCB status.
Today's an interesting day to launch this because the FCC is voting to ban all 126 test labs in China and Hong Kong. Not just the government-controlled ones they banned last year - all of them. 21% of the global total gone. 27 of those 126 are Western firms (Intertek, SGS, TUV, UL) operating China offices. I wrote up the full impact analysis at markready.io/blog/fcc-bad-labs-vote.
Full disclosure: I've never actually gotten a device through FCC certification myself. I've been building RF hardware since I was a kid but always on the hobby side. What pulled me into this was the data problem - the FCC publishes all this information but nobody had stitched it together into anything usable, and it felt like a genuinely interesting dataset to enrich and a real gap in understanding the hardware product space.
Built with Next.js + Cloudflare. The enriched dataset covers 28 countries.
> Today's an interesting day to launch this because the FCC is voting to ban all 126 test labs in China and Hong Kong
Does this increase or decrease demand for your tool? Less fragmentation would be expected to decrease demand from insiders. But more regulatory scrutiny would raise the stakes for outsiders getting it wrong.
I think this increases the relevancy for these tools and information. Gone will be the days of just sending your design to manufacturer in China and having it get fully certified and built through just one contact.
[flagged]
At this point I wish it were against the rules to accuse people or complain about articles as written by LLMs. It's creating so much noise that useful commentary is hard to find.
I don't see any signs of the parent comment being written by an LLM other than it's detailed and well-written.
Their comment does not give off "LLM-written" really... It drives forwards actual points without superfluous segments. I don't think it's helpful to try and discredit people whenever we want by throwing around accusations of "LLM-written".
That uncomfortable feeling when someone calls you an LLM...
This doesn't come across as LLMish at all to me.
*guidelines