Very cool! Can you reveal how much this custom board cost you?

For five fully assembled boards I paid 130 Euros, ordered the E-ink displays from a local shop

I built a project board nearly identical to this one a few years ago, and this seems about right. I don't think the displays themselves are included in the 130 euros. They can vary a lot in price depending on spec and supplier.

I built a project nearly identical to this one so knowing the level of complexity required, I'm not surprised it worked. AI is indeed raising the floor. Even simple hobby designs were once out of reach, but now we can "YOLO it and build stuff fast without necessarily knowing about the details" as the author puts it.

That said, without experience it's easy to pierce the veil of "simple circuitry" by accident, just like in XKCD: Tasks[1].

Looking at preview of the author's next project at the end of the article, I'm not so sure the same strategy will work. The board is probably two orders of magnitude more complex. Without knowing the circuit details, I see many red flags (randomly placed ics and decoupling, likely wrong length matching, random thick traces indicating lack of foresight on power distribution, backwards connectors, the probably unnecessary use of BGA footprints, etc).

[1]: https://xkcd.com/1425/

As someone who does both hardware design (mostly high precision analog electronics) and software engineering, the difference between both fields in AI capability is striking. Models like Fable and Astra can provide lots of interesting insights into the design of specialty electronics when prompted correctly. Yet they can hardly put that knowledge to work if instructed to generate a schematic or lay out a PCB.

Of course, for readers here with a solid grasp on how LLM's operate it's not exactly surprising. Still, I feel this is exactly the kind of thing that drives home that the way we reason about (general) intelligence does not map cleanly to LLM operation, no matter how advanced the current models are.

If anything, it is surprising that the LLM is so bad at stealing a ready-made RP2350 design! There are plenty of easy-to-find tutorials like [0] and [1] that handhold you through the entire design process, and a simple search will give you dozens of design files for open-source boards.

At this level of board complexity you almost have to try to make it completely non-functional, so although it is nice to see it in practice I am not exactly surprised that a LLM with a decent bunch of assistance managed to do it. But there's quite a gap between the equivalent of poorly copying a "Hello World" from StackOverflow and making actual production-ready boards.

[0]: https://pip-assets.raspberrypi.com/categories/1214-rp2350/do...

[1]: https://jlcpcb.com/blog/how-to-design-layout-with-rp2350

You have to be more specific about what you mean by design. The models are excellent and the netlist and the components along with tradeoffs.

Maybe it's not properly benchmarked.

A standard benchmark using kicad to create or fix electronic designs would go a long way in motivating and measuring improvement.

There's eebench, but maybe it's not exactly what you were thinking

i noticed the same treying to just generate 3D models using AI, i wonder if a well trained very specific world model would be better for the task

Me too, LLMs are bad at generating 3D models still but that and EE/ME CAD are the next frontiers.

Yeah enjoy it while it lasts. This is basically where LLMs were with coding a year or two ago. Give it a couple more years...

I'm still waiting on my order of 15 PCBs done with Sol - a mini-BT-keyboard that I hope will be comparable to the Psion 5 / Gemini PDA ones, using Kailh PG1316S switches. I still need to design and manufacture a housing (I do have a lot of hobbyist experience with this though), but the idea is to have a magsafe-connectable keyboard I can clip onto my Pixel and make use of the new Linux subsystem (or Termux) and do some productive stuff on the go. It's just been manufactured by JLC but not shipped yet. Funny enough, production itself was ~$70, then $80 for shipping, taxes and fees (tarriffs?).

Instead of directly generating Kicad, I chose to let it write "generators" in Python that output Kicad (S-Exps!) pcb and project - with mixed results - but I can generate a version with/without per-key Neopixels and customize certain parts of the board, including debug-stuff. I also _think_ this makes it easier to resume working on this in new sessions but not sure. Otoh, I never actually looked at the Python (vibed...) so it's still a lot of hard-coding.

I also landed on and used KicadRoutingTools after trying FreeRouting, combined with some minimal "auto-placing" code for larger ICs and such that are placed with certain constraints. My routing runs are way slower though - on the order of minutes.

I always wanted to try atopile - that might be a good match for AI/PCB work. The only thing that really held me back was the lack of routing.

I always wanted to try atopile - that might be a good match for AI/PCB work.

Unfortunately all public open source work halted 6 months ago and they pivoted to a browser based IDE product to monetize: https://atopile.io/blog/atopile-v16

AI is exceptionally good at RP2040/RP2350/ESP32-based boards, we've ordered ~15 unique designs, mostly RP2040-based that were designed and autorouted with AI using tscircuit and usually OpenAI models.

Currently the major mistakes that AI makes: 1) Improperly rotated connectors (backwards USB port) 2) Endless routing loops for complex boards 3) Really huge/unreadable single page schematics

I'm extremely bullish on AI for PCB design, but there is a huge, huge tooling gap. We basically need to soup up all our collective design rule checks so that AI can work in corrective loops. It needs to check for things like "opamp schematic layout conventional" and "trace width closely matches reference design for chip". One thing that makes me super excited is supply-chain resiliency when chip/subcircuit-swapping becomes super easy

Bought to us by Fable 5 and the more-experienced colleague who foresaw potential issues.

If it's functional: that's a breakthrough.

But: the PCB itself is not looking beautiful in terms of routing and a lot of schematic details yet. Horrible to to be more honest. But that's a minor issue.

I'm not overly shocked. 90% of creating PCBs is copying reference schematics from vendor datasheets and connecting them. That's LLM territory. This mostly failed for now as they just did pdf2text and killed all documents.

The last 10% are brutal (HF, EMC, high frequency multi-layer designs, etc.). But I managed LLMs to automatically create ngspice models using vision onto datasheets.. lol

Shameless plug for the slop coded CLI I use in my workflow: https://github.com/akiselev/datasheet-cli

It just sends the PDF to Gemini Flash and asks it to extract semi structured data (semi structured in the sense that it's not fully structured extraction, but its a JSON file to feed into the next step of the LLM design process).

Works pretty well in general, though I should update to the latest Gemini version.

That's pretty cool! How reliable has the footprint extraction been?

Upon Claude's compilation of the project, I asked it to prepare the required files for JLCPCB, and tell me what to select on their GUI

It would be an interesting step to get Claude (or other LLM) to do order the PCBs itself on the JLCPCB website.

We're nearly at the stage where you can describe what a device should do, let it design the circuit (using available components on JLCPCB), design a PCB and enclosure then order PCBs, assembly and 3d printed enclosure and get it sent to you for final assembly.

I've been working on several PCB projects, and agents are very helpful for automating many parts of the workflow. They're excellent for schematics. Complex routing, though, is still a human's job for now; I've watched Freerouting fall on its face many times. If you're doing six-layer high-speed DDR routing, AI won't solve that for you today.

However checkout EDA tools https://github.com/l3wi/claude-eda/ https://github.com/drandyhaas/KiCadRoutingTools

Yeah I've built about a dozen boards using GPT-sol 5.6 and my favorite is my Pi CM5 carrier board, to replace a laptop board in an old Sony Vaio p laptop. https://x.com/kanwisher/status/2089629411364745351?s=20

In our company, we are doing high speed PCB designs with FPGAs, memories, ADCs, DACs etc. We are not yet using AI much in our PCB design workflow because our designs seem to be too complex for the AI solutions we have tried. Are there any areas where people are successfully incorporating AI to their PCB design workflow, even with complex, high speed designs? Schematic/layout review? Routing? Footprint creation?

Why wouldn't an LLM be able to do high speed designs? Are you inventing new types of design rules for every board? Or is it simply because the automatic layout tools are not up to the task yet?

Yeah, we're still very far from a machine routing PCBs. Nice project, though, looks pretty fun to do it with an AI.

I researched a bit on how to manufacture PCBs at home, and i came across Voltera V one. The price is indeed way too high, but I am hoping somebody will make a foss version of it. There are already pretty interesting foss PnP machines.

For a 2-layer pcb I would either go chemical etching or a small routing cnc. Chemical etching kits , uv lamps and blank pcbs are not "too" expensive, but it's more of a professional setup than a hobbyist. Just think of all the PCBs yoi can send to manufacture (and with 4 layers and more) for the price of even getting half the stuff.

real PCBs are so much better than anything you can do at home, and so cheap, and so easy to order now, it's just not worth it

That said, I was not feeling much of an accomplishment, rightfully so. I used to enjoy the learning and the struggle that came with it.

Yes it does suck that the joy we had while building has been sucked out of us

For my hobby projects tho, I will continue to YOLO it and build stuff fast without necessarily knowing about the details.

I don't understand this conclusion. LLMs sucked the joy out of it, didn't feel rewarding, you're forced to use them at work, so you decide... just keep doing it at home, too? Why not actually learn to build a proper PCB knowing you can do much better than this horrible (despite being dead-simple) board that only got there with the help of your expert colleague's guidance anyways? Don't succumb to utter laziness and deny yourself the satisfaction of learning new skills, that's letting the thing rot your brain out worse than any TV ever did.

You are right. What I meant in my conclusion was more about : "I will use these tools to generate parts of a project I don't necessarily care about." I already started learning about proper PCB design, and I am actively fighting against choosing an LLM instead of grasping a concept fully. It's something I am struggling with, but all this is new for most of us. Hoping I will find a good balance at the end. Thanks for your remarks :)

Fair enough. I think there is great value to using LLMs to assist in learning new hobbies, helping lessen the friction, and doing research. I just think it's sad if you let them replace your thinking completely instead of augmenting it. Cheers for the human-written blog post, I did enjoy reading it.

I saw the board cost, but how about the total cost in tokens to design it? Can empathise with the good and meh feeling you got from it. For me I try to stick with manual struggles with new concepts which grows my brain and allows me learn something new. But its impressive what LLMs can do.

I have used fable to aid me with PCB design. It is pretty good. But it can also sometimes make very human mistakes. It mixed up RX and TX on a uart line. If you don't catch stuff like that and get assembled boards this can obviously be pretty annoying. On the other hand I have also mixed up RX and TX before, lol.

Hey - I ordered a board yesterday. I used Opus 5, Astra, and Sol at points to QC the board, and offer power routing advice. (The MCU is a brand new one, and poorly documented unless they send you an NDA and the full docs).

Those models were not capable of even schematic editing or building. (I didn't bother with routing or placement). They made gross errors all over the schematic; changed the known-working parts and net routing I'd copy+pasted from previous boards, changed I2C to SPI for no reason, and generally produced an unusable, unreadable result. Overlapping symbols and labels etc.

What do you mean by power routing for an MCU? Like they wanted to fatten the traces or what?

What PCB software did you give it? I'm having great success with Fusion (formerly known as Eagle) and others seem to be doing great with KiCad.

It honestly just makes me think there’s little reason for me to be alive anymore.

I hope this is a joke and you really don't consider this as such.

It's in fact opening up too many opportunities to us humans. It's never been this easy to acquire information and accomplish things.

It’s not a joke. My goals in life were all based around developing mastery for a meaningful purpose, and now there’s no need. What keeps me going is that at least I have some consumerist goals like understanding consciousness and physics or mathematics.

Computers can code for you, but it's not a reason to forget about your mastery goals.

The printing has been around for ages now, still people master calligraphy.

Yes! I have made and ordered 3 PCBs with Astra and Autodesk Fusion since the Astra release. First one has arrived and worked great. I especially love how the model can wrangle all the JLC stuff. What a time we live in