October is Open Hardware Month!

Tindie and the Open Source Hardware Association (OSHWA) have been friends for a long time! Every year, we like to support them during Open Hardware Month. We encourage our sellers to get their projects/products certified as open hardware. The process is not difficult, and the application on their website walks you through the process! Once you’re certified, this can be highlighted on your product listing on Tindie to let people know it’s certified open hardware.

This year, OSHWA are once again doing a 24hr livestream showing off the coolest open hardware over the last year. It will be on October 10th, starting at 8AM EST and ending 24 hours later. The theme this year is “good slop” – they want to see your messy, sloppy hardware prototypes that are 100% human created. In their words: “… we are looking for your work-in-progress projects, things that are maybe a little silly, projects you have invested time into that not a ton of people will understand why, passion projects, side quests, anything that really took you down a path you didn’t expect, projects that use bioplastics, goop, mycelium, slime, or any other strangely viscous substance!”

If you’re interested in participating in the livestream, you can sign up for a 20-40 minute slot to show off your latest work(s) and talk about open hardware with the community. We encourage those who can to support the amazing work OSHWA does by getting a membership to OSHWA – they rely on members to do all the cool stuff they do every year, like the Open Hardware Summit among many other initiatives.

Throughout Open Hardware Month, we’ll be highlighting open source hardware available on Tindie. If you’ve got an awesome project that hasn’t been covered before and is certified open hardware, get in touch via our social media or email and let us know about it so we can write a blog post about it!

source https://blog.tindie.com/2026/09/october-is-open-hardware-month/

Clock Module: Homebrew SBC Tool

I’ve played around with designing my own microprocessor-based single-board computers over the years. I was inspired originally by a Z80 project done by a rather unknown YouTuber, and then was very inspired after seeing Matt Sarnoff’s Ultim809 project. I always wanted a computer with a front panel full of LEDs and switches so I could program, quite literally, by hand. When you first start these types of projects, the most difficult part can be figuring out all the logic for memory mapping and I/O control, and if you don’t have a very wide logic analyzer like an HP 1650B then you need a way to single-step the clock and probe the buses. And that’s where the Clock Module DIP comes in!

It’s a beautifully simple module, with a 16MHz crystal and a 74HC4060 oscillator/divider which outputs 11 different frequency taps, from 16MHz all the way down to 976.5625Hz. There is also the all-important debounced single-step clock button, which is a must for static CPUs that can have their clocks stopped without issue (like the Z80). The other IC is a hex inverter in case you want the opposite phase clock output — this can be very useful for some microprocessors that require multi-phase clock outputs, though some need a 90° phase offset instead of 180° that the 74HC14 gives.

The best part is it’s fully through-hole, easy to assemble, and uses a reliable and easy-to-understand circuit. There’s even an SMD option if you want a smaller board! And, if you don’t want to bother with assembly, you can order it pre-assembled too. So the next time you’re breadboarding your 6809, Z80, 6800, or 6502, pick up a Clock Module from spartatux!

source https://blog.tindie.com/2026/09/clock-module-homebrew-sbc-tool/

When Large Models Start Designing PCBs, What Happens to the Electronics Industry?

I’ve been using large language models to design a few circuit boards recently, and the more I use them, the more I feel that something important is starting to change.

Until recently, when we talked about AI-assisted hardware design, we mostly meant asking AI to search for information, recommend components, explain a datasheet, or write some firmware. But once it came time to draw the schematic and lay out the PCB, an engineer still had to open KiCad, Altium, or another EDA tool and do the work manually.

That boundary is beginning to move.

Large models can now understand KiCad project files, modify schematics, place components, route PCBs, run DRC, and sometimes correct their own mistakes after checking the result.

They are nowhere near reliable enough to design every complex product on their own. But the direction is becoming clear.

And I don’t think this will only affect the EDA industry.

It could reshape a much larger part of the electronics business, from design tools and component databases to manufacturing, test, supply chain, and even the way hardware products are sold.

The first thing to lose value may be the skill of operating an EDA tool

Building an electronic product has traditionally had a high barrier to entry.

You need to understand analog and digital electronics. You need to choose components, read datasheets, draw schematics, and then spend a long time learning how to use Altium, Cadence, KiCad, or some other EDA package.

A lot of people with good product ideas never get past that point.

Something similar has already happened in software.

A few years ago, if you couldn’t program, you simply couldn’t build software. Today a product manager, designer, student, or hobbyist can use tools such as Claude or Codex to create a surprisingly capable application.

Hardware may be heading in the same direction.

Imagine a mechanical engineer, industrial designer, or maker saying:

I want to build a small robot that uses a camera to recognize and sort objects. It should use an ESP32, drive a few motors, include a microphone, and run from USB-C.

The model could choose the major components, find reference designs, build the schematic, lay out the PCB, generate the BOM, and prepare the manufacturing files.

At that point, “knowing how to use Altium” is no longer such a high barrier.

That does not mean EDA tools disappear.

In fact, I think many EDA tools may start to look more like compilers in software development.

Programmers can use AI to write code, but that does not make compilers, tests, or debuggers less important. If anything, the more code gets generated, the more important verification becomes.

PCB design is similar.

A model can generate a board, but someone—or something—still needs to determine whether the design is electrically correct, whether the power system is stable, whether the signal integrity is acceptable, whether it will overheat, and whether it can pass EMC.

Those questions depend on the physics engines, simulation tools, design rules, and verification systems behind the EDA interface.

So I expect engineers to spend less time manually clicking through complicated menus, while AI agents increasingly call these tools in the background.

The value of the UI may decline.

The value of the solver, router, constraint engine, SI/PI analysis, thermal simulation, and physical verification may increase.

The large companies are already moving

It is interesting to look at what the major engineering software companies have been buying.

Synopsys spent roughly $35 billion to acquire Ansys.

Siemens spent around $10 billion to acquire Altair.

Cadence has continued expanding its simulation portfolio and is introducing agent-based systems for PCB and advanced-package design.

Renesas is even more interesting.

A semiconductor company best known for MCUs, analog, power, and connectivity products spent A$9.1 billion acquiring Altium.

These companies clearly do not believe EDA is becoming irrelevant.

They are expanding EDA in two directions at the same time:

toward AI, and toward the physical world.

Traditional EDA mainly answers:

How do we create this electronic design?

The next generation of engineering platforms may need to answer:

How do we turn an idea into a verified product that can actually be manufactured?

That is a much bigger problem.

And it means the boundaries between EDA, simulation, PLM, component data, manufacturing, and supply chain will become less clear.

Component databases may become much more valuable

Today, an engineer searching for a component usually visits DigiKey, Mouser, LCSC, Arrow, or another distributor and manually filters the results.

In the future, the engineer may not do the search at all.

An AI agent could simply ask for:

• 5V to 12V input
• 3.3V / 2A output
• efficiency above 90%
• cost below $1
• production lifetime of at least five years
• at least tens of thousands of units currently available

Then it could select the part itself.

When that happens, the appearance of a component supplier’s website becomes less important than the quality of the data behind it.

The agent needs accurate information about:

• electrical parameters
• pin definitions
• package dimensions
• symbols and footprints
• 3D models
• SPICE and IBIS models
• reference designs
• alternates
• lifecycle status
• inventory
• pricing
• application limitations

In the past, this information was mainly prepared for engineers to read.

In the future, much of it will also need to be prepared for machines to understand.

DigiKey, Mouser, Arrow, Avnet, LCSC, and other distributors may eventually compete not only for engineers, but for the agents working on behalf of those engineers.

The supplier with the best structured data, the most useful API, and the most accurate real-time inventory may have a better chance of being selected automatically.

The same applies to semiconductor companies.

TI, ADI, ST, Renesas, Infineon, and others may increasingly have to think about a new question:

How do we make it easier for an AI system to design our devices into a product?

That is one reason I find the Renesas acquisition of Altium so interesting.

It suggests a semiconductor company can move from simply selling chips to helping customers design complete systems around those chips.

PLM may become more important, not less

PLM has traditionally been something that becomes important after the design has already been created.

It manages files, BOMs, versions, changes, approvals, and product history.

But once AI agents start doing more engineering work, PLM may become even more important.

The biggest problem for an engineering agent is often not drawing the schematic.

It is understanding the context of the project.

Why did the previous revision use this component?

Why was it replaced?

What did the customer ask for?

Which component caused problems during the last prototype?

What changed between revision 17 and revision 18?

Which parts have already been approved by the company?

What waveform should we expect at TP7?

Most of that information is not inside a schematic file.

It is scattered across PLM systems, ERP systems, test reports, purchasing records, emails, and the memories of engineers.

That is why I increasingly think of EDA as the tool used by the agent, while PLM may become the long-term memory of the agent.

We have been running into exactly this problem while adding online KiCad viewing, PCB inspection, iBOM, BOM processing, and various agents to ezPLM.

It does not make much sense to ask a large model to reread an entire KiCad project every time someone asks a question.

A better approach is to convert the schematic, PCB, components, nets, BOM, test points, and other design information into a common hardware design graph.

What the AI really needs to know is something like:

U7 pin 3 connects to R23, which connects to the ADC input. That net is routed for 18 mm on the PCB with no vias, and a high-speed clock passes 3 mm away.

It does not need thousands of UUIDs, font settings, drawing coordinates, and other details that exist only to operate the CAD tool.

I think this kind of structured hardware data may become an important part of the infrastructure for AI-assisted electronics.

PCB manufacturers may be some of the biggest winners

A common reaction is:

If AI can design PCBs using KiCad, what happens to companies that built an ecosystem around their own EDA tools?

I actually think manufacturing platforms such as JLCPCB could benefit from the change.

Their real advantage is no longer only the EDA tool.

They have connected:

EDA, components, PCB fabrication, assembly, CNC machining, and 3D printing.

JLC’s business has grown far beyond being a simple PCB fabrication service.

If AI lowers the barrier to hardware design and increases the number of people capable of creating PCBs by ten times, that could mean more prototypes, more component orders, more SMT assembly, more CNC parts, and more 3D-printed enclosures.

The part that may change is the lock-in created by the design tool.

Today, if someone spends years learning a particular EDA platform, there is a natural tendency to stay inside the surrounding manufacturing ecosystem.

Tomorrow, an AI agent may generate a KiCad project and automatically ask several suppliers:

How much will this board cost?

When can you deliver it?

Can you source every component?

What is the expected yield?

At that point, the real competitive advantages of a manufacturer become:

price, lead time, quality, component availability, DFM capability, logistics, and APIs.

The manufacturer that is easiest for an agent to work with may have a better chance of winning the order.

The faster AI generates designs, the more valuable testing becomes

This is one of the changes I think is still underestimated.

In the past, an engineer might spend two weeks creating one PCB revision.

In the future, AI could generate ten—or even one hundred—candidate designs in the same period.

Then the main question becomes:

Which one actually works?

As design generation becomes cheaper, verification becomes more valuable.

This is one reason simulation companies have become attractive acquisition targets, and why the major EDA vendors continue adding more physics to their platforms.

The same thing will happen in the real world.

Once the PCB arrives, somebody still needs to answer questions such as:

Is the power ripple acceptable?

Why does USB enumeration occasionally fail?

Why is there noise in the camera image?

Does the measured waveform match the simulation?

Eventually, AI will need to connect the design in the computer with the physical board sitting on the bench.

That means oscilloscopes, logic analyzers, multimeters, cameras, and automated test equipment may become the eyes and hands of the engineering agent.

The AI designs a PCB, sends it to manufacturing, creates a test plan, measures the finished board, compares the result with the expected behavior, finds the problem, and then modifies the next revision.

That closes the loop:

Design → Manufacture → Measure → Analyze → Modify

When that loop becomes practical, hardware development changes much more fundamentally than it does when AI simply learns to draw a schematic.

Modules probably will not disappear

If more people without deep electronics backgrounds begin building hardware, what will they use first?

Probably modules.

ESP32 modules, camera modules, motor drivers, power modules, and sensor modules are much easier for an agent to combine than designing everything from transistors, resistors, and capacitors.

So I actually expect the module market to grow in the short term.

But the definition of a good module may change.

Today, a module often comes with a PDF datasheet and a few examples.

In the future, a really useful module may need to provide:

• schematic symbol
• footprint
• 3D model
• reference schematic
• firmware
• driver
• simulation model
• test procedure
• operating limits
• known-good measurements

Ideally, it should also describe its behavior in a form an agent can understand.

At that point, a module begins to look less like a traditional breakout board and more like an npm package in software.

The same may happen to reusable circuits.

A proven USB-C PD power stage, a 10 MHz analog front end, or a BLDC motor driver could become a kind of hardware IP block that an agent can search for, instantiate, parameterize, and verify.

That could create an entirely new market for reusable, verified hardware IP.

Eventually, the bottleneck may become finding customers

When hardware gets easier to build, the difficult question may shift from:

Can I make this?

to:

Who will buy it?

The software industry is already there.

Using AI to build another app is becoming easy. Getting people to use it is still hard.

Hardware may follow the same path.

We may see many more small teams, makers, and even individuals creating their own products.

That could make platforms such as Tindie, Crowd Supply, and other specialized marketplaces more valuable.

If the number of new hardware products grows dramatically, scarcity moves further down the chain.

Manufacturing capacity matters.

Quality matters.

Trust matters.

Reviews matter.

Community matters.

Distribution matters.

Getting discovered matters.

That is why I do not think the most valuable company in the future electronics industry will necessarily be the company with the best PCB editor.

The bigger opportunity may belong to whoever can connect the entire path:

Idea → Design → Components → Simulation → Manufacturing → Test → Sell

Large models can reduce the barrier at the front of that chain.

But somebody still has to turn the idea into a reliable physical product and then get that product into the hands of a customer.

So I do not think AI will make the electronics industry smaller.

Quite the opposite.

The software industry became enormous not because software became harder to build, but because it became easier.

If millions more people become capable of turning their ideas into electronic products, the markets for PCBs, components, manufacturing, testing, modules, reusable IP, and sales platforms may all become much larger.

What loses value is one particular skill:

knowing how to operate a specific tool.

What gains value is very different:

verified design knowledge, trustworthy data, physical simulation, manufacturing capability, test data, and access to customers.

If this transformation really happens, the largest electronics companies of the future may not fit neatly into categories such as “EDA company,” “PCB manufacturer,” or “component distributor.”

They may look more like a combination of:

GitHub + npm + AWS + Shopify — but for hardware.

Whoever can connect an idea to a real product that people can actually buy may have the biggest opportunity of all.

source https://blog.tindie.com/2026/09/when-large-models-start-designing-pcbs-what-happens-to-the-electronics-industry/

I Have an Idea #01 – What If Your Car Could Remember the View You Just Saw?

A recent road trip gave me a new product idea.

Before the trip, I bought an Insta360 camera because road trips are full of moments you cannot plan for: a mountain suddenly appearing around a bend, a lake opening up beside the highway, a sunset over the coast, or a deer crossing the road for just a few seconds.

A 360° camera sounds perfect for this. In practice, I found it surprisingly awkward.

• If I mount the camera inside the car, I have to deal with windshield reflections, dashboard glare, dirt, rain and limited angles.
• If I mount it outside, I need a reliable bracket, weather protection, power, and some confidence that it will still be there after two hours on the highway.

Then I discovered the bigger problem:

I am driving.

When something beautiful appears, I should be looking at the road, not searching for a Record button.

And by the time I think, “I should record this,” the best moment may already be behind me.

That made me wonder – Why should the driver decide when to start recording?

Modern cars already have plenty of cameras.

---

Cars Already Have Eyes

A modern vehicle may have a forward ADAS camera, surround-view cameras, parking cameras, rear cameras and digital mirror cameras.

Some automakers are already using those cameras for recording.

• BMW Drive Recorder can use the vehicle’s surround-view cameras to capture driving footage, including scenic routes.
• Polestar 4 can use an exterior-facing camera for photos and video, and its dashcam can be controlled by voice.
• Garmin dashcams already let drivers say things like “Save Video” and preserve footage from around the time of the command.

So the building blocks are already here.

But I think there is still a missing step. Most automotive cameras were designed for machines, not for beautiful travel video. ADAS cameras care about lanes, pedestrians and traffic lights. Surround-view cameras often use extreme wide-angle or fisheye lenses because their job is to see the pavement around the vehicle.

That is very different from filming mountains, lakes and coastlines.

---

The Idea: RoadScape AI

I call the idea RoadScape AI.

I do not think of it as another dashcam.

I think of it as:

Automotive Visual Memory

The most important command would not be:

“Start recording.”

It would be:

“Save what I just saw.”

Imagine driving through the Canadian Rockies. A beautiful mountain appears on your right.

Ten seconds later, you say:

“Hey RoadScape, save that mountain on my right.”

RoadScape replies:

“Got it.”

A few seconds later, a clean video clip of that mountain appears on your phone.

You never touched the camera, you never touched your phone, and you did not have to predict the moment before it happened.

---

How Can It Save Something That Already Happened?

The cameras would continuously record into a circular buffer.

For example, they might always keep only the most recent five minutes.

NOW
│
├── Front Camera ── last 5 minutes
├── Left Camera  ── last 5 minutes
└── Right Camera ── last 5 minutes

Old video is overwritten unless the system decides it is worth keeping.

But RoadScape would go beyond a simple “Save Video” command.

Consider:

“Save the mountain I just saw on my right.”

That sentence contains three useful clues:

“just saw” — time
“on my right” — direction
“mountain” — meaning

RoadScape could search recent footage from multiple cameras, find the mountain, compare the available views and save the best one.

Now the system is not just recording video.

It is searching memory.

---

Three Cameras Instead of a Gimbal

If I built the first version, I would probably use three fixed cameras:

Left + Front + Right

The front camera would look straight ahead.

The left and right cameras would point perhaps 60–70 degrees outward.

Together, they could cover most of the useful scenery in front and beside the vehicle.

I would avoid a mechanical gimbal.

• No motors.
• No moving parts.
• No waiting for the camera to rotate.

The software turns its attention instead of the hardware turning its head.

I would also mount the camera outside the windshield, ideally in a low-profile bar on the roof or roof rack. That avoids many of the reflections and obstructions that made me start thinking about this idea in the first place.

---

A Maker Could Build This Today

This is where I think the idea becomes especially interesting for the Tindie community.

The first prototype does not need a custom PCB.

It could be built from existing modules:

• three USB cameras,
• a Jetson Orin Nano, RK3588 board or even a laptop,
• an SSD,
• a GNSS module,
• an IMU,
• Wi-Fi,
• a 12V power converter,
• and a 3D-printed camera mount.

The first version does not even need 4K. 1080p is enough to answer one question:

Does “Save What I Saw” actually work?

The software can also be built step by step.

First:

“Save that.”

Save video from a few seconds before and after the command.

Then:

“Save the view on my right.”

Now add direction.

Then:

“Save that mountain on my right.”

Now add visual recognition.

Later:

“Did we capture the deer we saw five minutes ago?”

Now it starts becoming true visual memory.

---

AI Changes How a Maker Can Build It

A few years ago, a project like this would have required a lot of software experience.

• Multi-camera video.
• FFmpeg or GStreamer.
• Speech recognition.
• GPS.
• Databases.
• Computer vision.
• A phone interface.

Today, an AI coding agent can help build those pieces one at a time.

For example:

“I have three USB cameras on Ubuntu. Keep the latest five minutes from each camera as rolling video segments.”

Then:

“Sample one frame every 500 ms and store image embeddings with timestamp and camera ID.”

Then:

“Build a mobile webpage with Front, Left and Right previews and a Save That button.”

A hardware Maker no longer needs to be an expert in every software layer before trying the idea.

The important skill becomes:

Can I break the product into the right problems?

That is a pretty big change.

---

From Weekend Prototype to Tindie Product

If the idea works on the bench, then I would start reducing the mess.

1. Replace the laptop or development board with a custom carrier board.
2. Replace USB cameras with embedded camera modules.
3. Add proper automotive power protection.
4. Add weather sealing.
5. Design a low-profile enclosure.
6. And instead of manufacturing thousands of units, I would probably make 20 or 30.
7. Put them on Tindie.
8. Let people who love road trips, RVs, overlanding, photography and computer vision try them.

The first product might not even need to be a finished consumer device.

It could be:

• a three-camera kit,
• a RoadScape development board,
• a printable mounting system,
• or a complete open-source developer kit.

That is also where the Maker community becomes useful.

• Someone may put it on a motorcycle.
• Someone may install it on an RV.
• Someone may use it on a boat.
• Someone may turn it into a wildlife observation system.

The product may become something different from what I originally imagined.

That is often the fun part.

---

And Maybe Crowdfunding Comes Later

I would not want to start with a few renderings and immediately launch a crowdfunding campaign.

I would rather do it in this order:

IDEA
 ↓
OFF-THE-SHELF PROTOTYPE
 ↓
WORKING SOFTWARE
 ↓
REAL DEMO
 ↓
20–50 TINDIE UNITS
 ↓
USER FEEDBACK
 ↓
CUSTOM HARDWARE
 ↓
CROWDFUNDING
 ↓
PRODUCT

Crowdfunding should help pay for tooling, custom electronics, weatherproofing, certification and manufacturing.

It should not be the first test of whether anyone wants the idea.

In the AI era, this may become a very interesting way for independent Makers to build products.

The old path was often:

Have an idea → find a team.

The new path may increasingly become:

Have an idea → buy a few modules → build the first version with AI.

Then ask:

Are there another 1,000 people who want one too?

---

Where Could RoadScape Eventually Go?

The most interesting long-term version may not need its own cameras at all.

Future cars may already have six, eight or ten cameras.

RoadScape could compare all available views and choose the one with the best sharpness, exposure, composition and least distortion.

The car already has the eyes.

RoadScape would provide the:

memory and intelligence.

At the end of a road trip, I do not want six hours of raw video.

I want to open an app and see:

287 km driven
32 scenic moments found
11 moments saved

Then ask:

“Make a one-minute video from the best mountain scenes today.”

Or:

“When did we first see the lake?”

Or:

“Did we capture that deer beside the road?”

At that point, RoadScape is no longer really a camera.

It becomes:

A visual memory of the journey.

---

What Do You Think?

I do not know if three cameras are the right answer.

Maybe one 360° camera is better.

Maybe cameras near the side mirrors would work better.

Maybe the real product should be software that uses cameras already built into the car.

That is exactly why I want to start this I Have an Idea series.

These are not finished products.

They may not even all be good ideas.

They are simply real-life problems that make me stop and wonder:

Could we build this differently today?

So, if you are a Tindie Seller:

How would you build RoadScape?

If you are a Buyer:

Would you put one on your car?

And if your vehicle already has a collection of cameras:

Would you like automakers to let Makers build new applications with them?

I would love to hear what you think.

source https://blog.tindie.com/2026/09/i-have-an-idea-01-what-if-your-car-could-remember-the-view-you-just-saw/

I Designed a KiCad PCB with GPT-5.6 and GPT-6 — and It Changed How I Think About Hardware Design

One of the things that surprised me most about the newly released GPT-6 Astra is how capable it has become at working with KiCad.

I happened to be planning a new hardware project at the same time: a small controller board based on Espressif’s new ESP32-S31 module, intended to become the “head” of a small desktop robot.

So I decided to use the project as a real-world test of GPT’s PCB design capabilities.

Before touching the electronics, I first asked GPT to generate a concept image of the robot from a simple description.

The PCB inside the robot’s head needed to include:

• An ESP32-S31 module as the main controller
• Camera, dual microphones, and a speaker
• An LCD display
• 12 RGB LEDs
• An IMU
• A ToF distance sensor

The input I gave GPT was surprisingly simple.

I uploaded two PDF schematics from Espressif reference boards based on the ESP32-S31, together with the robot concept image, and asked GPT to design the electronics for the robot head.

That was basically it.

First Test: GPT-5.6 in Chat

My GPT Work usage quota had just run out, so while waiting for the next usage window, I tried the same experiment using regular Chat with GPT-5.6.

For most of my software development work I currently use Claude Code, while keeping GPT running alongside it for architecture review, debugging, and alternative approaches. It is actually quite interesting to let two strong models challenge each other’s ideas.

The first thing I asked GPT-5.6 to generate was the schematic.

The result was technically interesting, but visually not very good.

The schematic was hard to read, and it was not consistently using the standard symbols from KiCad’s official libraries.

That said, I have been arguing for a while that traditional schematics may eventually become less important in much the same way that programming languages are gradually becoming less visible to the user.

What matters most to me is not how beautiful the schematic looks.

What matters is whether the electrical relationships between pins, nets, and functional blocks are correct.

So I asked GPT to redraw the power section using standard KiCad schematic symbols and conventional schematic formatting.

The result looked much more like something drawn by an experienced engineer:

I manually checked the connectivity and could not find any obvious errors in that section.

That impressed me.

Remember, I had only provided GPT with two PDF schematics. The ESP32-S31 was still very new, with relatively little reference material available online, and almost every component used in the generated design came directly from those two reference documents.

That tells me something important:

GPT-5.6 is no longer simply “drawing schematics.” It is starting to understand enough of the circuit structure to reconstruct and adapt an electronic design from reference material.

From Schematic to Component Placement

Next, I asked GPT to propose a PCB component layout.

Not bad at all.

The placement was logically organized around the product architecture rather than simply arranging components mechanically.

In Chat mode, however, many of the components did not appear to have their corresponding 3D models associated with them.

Then things became much more interesting.

GPT-6 Astra in Work Mode

Once Work became available again, I submitted essentially the same requirements to GPT-6 Astra.

After roughly three iterations, it produced a complete compressed KiCad project that could be opened with KiCad 7 or newer.

After extracting the archive, I found much more than just a .kicad_pro, schematic, and PCB file.

GPT had generated a fairly complete engineering project structure with a number of supporting files.

The schematic used a hierarchical design structure, divided into 13 functional blocks.

The components on each sheet were spaced rather generously, but every page represented a clearly separated functional module.

More importantly, the schematic quality was dramatically better than my earlier Chat experiment.

The design consistently used symbols from KiCad’s official libraries, and the net labels were placed in a clean and conventional way.

It was actually pleasant to read.

Then I opened the PCB.

Apart from nine unrouted nets, the component placement and routing were surprisingly reasonable.

And when I opened KiCad’s 3D Viewer, most of the components already had 3D models attached.

Again, these were largely models from KiCad’s standard libraries.

Is It Ready to Manufacture?

Not yet.

There are still things I would definitely review before sending this board to fabrication.

For example, I have not yet completed a detailed review of every schematic block, and the antenna keep-out and placement around the ESP32-S31 module still need improvement.

There are also some layout decisions that I would change manually.

But that is almost beside the point.

The remarkable part is that GPT created something this complete from only:

two PDF reference schematics + a short product description + a concept image.

No carefully prepared netlist.

No existing KiCad project.

No detailed component-placement instructions.

No step-by-step guidance.

It simply worked through the problem and produced a usable engineering starting point.

If I continue from this generated project and perform the normal engineering review, corrections, and optimization, I estimate that it can already save me well over half of the time I would normally spend on the initial design.

The More Interesting Question: What Happens When the Loop Closes?

Today’s workflow is still essentially open-loop.

The model reads reference material, reasons about the design, generates a schematic and PCB, and then hands the result to the engineer.

But imagine connecting that same model to a complete verification loop:

• Circuit simulation
• ERC
• DRC
• Signal-integrity checks
• Power-integrity analysis
• Thermal analysis
• DFM checks
• Component availability
• BOM cost optimization
• Datasheet verification
• 3D mechanical checking

Now the model would no longer just generate a design.

It could generate, test, diagnose, revise, and verify the design repeatedly until it satisfied a set of engineering constraints.

That changes the nature of PCB design quite dramatically.

A senior hardware engineer might spend several days turning an idea into a reasonably mature first PCB revision.

I can now imagine AI-assisted workflows producing a comparable starting point in well under an hour for many conventional embedded designs — with the engineer spending most of the time reviewing decisions rather than manually creating every symbol, net, footprint, and trace.

And as these verification loops improve, the quality gap will shrink quickly.

Maybe Schematics Will Become Like Source Code

I have held a slightly unpopular view for some time:

traditional schematics may eventually become an intermediate representation rather than the primary interface for hardware design.

Something similar is already happening in software.

Developers increasingly describe intent in natural language while AI systems generate Python, JavaScript, SQL, configuration files, and infrastructure definitions underneath.

We still need the code.

But humans do not necessarily need to write every line of it.

PCB design may follow the same path.

Future hardware engineers may spend much less time manually placing symbols and routing nets, and much more time specifying:

• Functional requirements
• Electrical constraints
• Mechanical constraints
• Cost targets
• Manufacturing constraints
• Reliability requirements
• Test requirements

The CAD files will still exist.

The schematic will still exist.

The PCB layout will still exist.

But they may increasingly become machine-generated engineering artifacts rather than the place where the design process begins.

For makers, this could be particularly powerful.

A huge amount of time in hardware prototyping today is spent not on the creative part of the project, but on repeatedly doing very familiar engineering work: selecting known circuit topologies, reading datasheets, creating footprints, wiring common interfaces, placing decoupling capacitors, checking pin mappings, and routing fairly conventional boards.

AI is getting very good at exactly this kind of work.

My ESP32-S31 robot head is still just an experiment.

I certainly would not send the current version directly to manufacturing without reviewing it.

But after seeing what GPT-5.6 and GPT-6 Astra can already do with KiCad, I think we are much closer to a different kind of hardware design workflow than many engineers realize.

Instead of opening KiCad and asking:

“Where should I start drawing?”

we may soon start with:

“Here is the product I want to build. Generate the first manufacturable version, show me the risks, and tell me what I need to decide.”

That is a much more interesting way to build hardware.

And I suspect we are going to see progress here much faster than most of us expect.

source https://blog.tindie.com/2026/09/i-designed-a-kicad-pcb-with-gpt-5-6-and-gpt-6-and-it-changed-how-i-think-about-hardware-design/

KeyVault32: Offline Password Manager

Editor’s note: Remember that security works in layers, and that we are not able to do a forensic analysis of this device to ensure there are no bugs in the code. However, the code is available which is a huge plus versus closed-source hardware!

Everyone should be using a password manager these days. The risk from using the same password across multiple sites is just too great — all it takes is one breach and suddenly you lose control of all your accounts. Cloud-based password managers are convenient and are in theory safe, but there have been some breaches. Having a hardware-based password manager like the KeyVault32 leaves your data in your hands, removing a potential threat vector.

As far as convenience goes, the KeyVault32 connects to any device that can accept keyboard input and can automatically input usernames and password for you. A built-in web server and on-board WiFi allow easy access to the entire vault for adding and updating entries. All passwords are encrypted using AES-256-GCM with key derivation coming via SHA-256 hashes, both of which are accepted and current methods of encrypting and securing data.

Another advantage is that you can use it anywhere, anytime – no requirement for a web connection (though many cloud password managers do have a local cache) and it doesn’t require any drivers or software as it just appears as a USB Human Interface Device which all operating systems will automatically recognize and use.

So, if you want a hardware storage option, even as a backup to your cloud vault, consider the KeyVault32.

 

source https://blog.tindie.com/2026/09/keyvault32-offline-password-manager/

Deadline Approaching: Tindie x NextPCB R&D Program

Just a reminder to all our sellers and readers that the Hardware Creator R&D Support Program application deadline is just a few days away – August 31st, 2026! You could get up to $150 in PCB prototyping support for your upcoming projects/products — so make sure to check out the official signup page for all the details!

Just a recap for those who may have missed it:

Each chosen seller will receive $150 worth of support over the three months for PCB prototyping and international shipping, and access to NextPCB’s HQ DFM tool. Eligibility is simple: you have to have a Tindie account with good standing that has at least one sale, with a few limited slots for new sellers who do not yet have any orders. Having a good product development plan, the ability to demonstrate innovation capabilities or a credible long-term operating plan will make your application stand out!

And we will also be specially highlighting projects under this program via our blog and social media to help stir up interest in the cool things you are working on. So apply before Monday is over and get support with your next project!

source https://blog.tindie.com/2026/08/deadline-approaching-tindie-x-nextpcb-rd-program/

Gotta Catch Em All! GameBoy Pokemon Trader

One of the unique features of the original Game Boy (and its immediate successors) was the ability to link two or more units together using the Link Cable. While this enabled multiplayer gaming for many games, no game used it as much as the original Pokemon games. Without a friend who owned the opposite colour game and a link cable, it was impossible to catch all 150 original Pokemon! Well, now you can trade anything you want with the Game Boy Pokemon Trader.

Of course, you could easily use this device to stack your team with custom Pokemon and completely obliterate the Elite Four (which carries a certain satisfaction if you, like me, struggled to beat them as a child), but it could also be an extremely useful tool for those researching how the Pokemon code works, how Link Cable transactions work, and content creators setting up custom scenarios (Soft Lock Picking, anyone?).

The interface is very easy to use – once powered up, connect to the WiFi network and browse to the IP printed on the custom PCB. Pick the traits you want for your traded Pokemon, connect it to your Game Boy with a Link Cable, and then visit a Pokemon Centre and enter the trading room. It’s just like trading with any other Game Boy, except you’re in total control of the entire process.

source https://blog.tindie.com/2026/08/gotta-catch-em-all-gameboy-pokemon-trader/

NextPCB x Tindie Hardware Creator R&D Support Program

I’m sure you’ve noticed on site by now, but we wanted to officially announce this new Hardware Creator Support program in partnership with NextPCB! We’re aiming to help our sellers to reduce development costs and help bring cool new products to the market. Eligible Tindie sellers can receive support for PCB prototyping, international shipping, and DFM design checks along with project promotion across our social media channels.

The program will run from September 5th, 2026 to December 5th, 2026.
Program coupons will be issued once per month on the 5th of each month for September, October and November. Each coupon is valid for three months from the date of issue.  The application deadline is August 31st so make sure to submit yourself before the cutoff date!

Each chosen seller will receive $150 worth of support over the three months for PCB prototyping and international shipping, and access to NextPCB’s HQ DFM tool. This tool is available both as a web tool and a desktop application and can check Gerber files, drill files, trace width/space issues, solder mask and silkscreen designs, and other common PCB manufacturing issues. It’s meant to work as an additional step on top of doing DRC in your CAD/CAM software, not as a replacement.

Eligibility is simple: you have to have a Tindie account with good standing that has at least one sale, with a few limited slots for new sellers who do not yet have any orders. Having a good product development plan, the ability to demonstrate innovation capabilities or a credible long-term operating plan will make your application stand out! Show us what your plans are for a new product, and give us a brief overview of how we can help bring it to life. You will also need to either have, or be able to register for, a NextPCB account. Ensure you fill out everything in the application form correctly, as accuracy is key to a successful application!

Additionally, we will also be doing project promotion support from September 5th to February 5th, 2027. There’s no need to apply for this separately: we will be covering many of the approved projects and we’ll work with you to tailor your messaging to help drive interest.

So what are you waiting for? Click here to read all the details and get started with your application today!

source https://blog.tindie.com/2026/08/nextpcb-x-tindie-hardware-creator-rd-support-program/

Secure TOTP Storage & Usage

Keeping online accounts secure is more important than ever, especially with breaches occurring all the time. Using a password manager is an important step, but if you want to take your security to the next level, it’s worth looking at secure hardware devices like the TOTPVault. This fully open-source project stores TOTP secrets and generates codes with them for 2-factor authentication (2FA), an extremely important security feature that prevents attackers from accessing accounts even with a breached password.

Separating the TOTP secrets from a general-purpose, online device like your smartphone significantly decreases the likelihood of your accounts being breached. Moving TOTP away from password managers is also important; while convenient, storing passwords and TOTP secrets in the same database degrades the security model provided for by 2FA.

The TOTPVault stores encrypted TOTP secrets on a RISC-V microcontroller, an ESP32-C3, and a different MCU handles the USB interface, separating the cryptographic processes from the data access methods. TOTP data are encrypted with AES256, and the decryption key is derived from a vault password using PBKDF2. JTAG has been permanently disabled, with eFuses set to prevent accessing the data inside the MCU. The WiFi/Bluetooth stack has been completely disabled, and there is no antenna on-board.

The project’s source code is fully available and therefore auditable; you can also build the firmware and flash it yourself to ensure the running code is exactly what you expect, reducing the amount you need to trust the seller. The ESP32-C3 was chosen because it has excellent hardware crypto support, good support for Rust, and it has been PSA-L1 certified. So if you want to take an easy step to increase your account security, especially if you work in a high-risk job like journalism, give the TOTPVault a look!

source https://blog.tindie.com/2026/07/secure-totp-storage-usage/

Give C64 Emulation a Realistic Touch

The Commodore 64 keyboard is one of the most unique features of that machine and its siblings the VIC-20 and C128. Some people love it, some people hate it – but it’s undeniably how millions of people interacted with the C64. While emulation has opened up a whole new era of enjoyment for vintage computers, some people miss the real feel of the keyboard and joysticks. Well, the KEY2USB makes this a cinch!

Simply plug the KEY2USB into the expansion port of the C64, connect it with USB to your modern PC, and use the keyboard and joysticks as USB HID devices! There is no modification of any kind needed to any part of the C64, so you don’t need to even open it up. It’s great for playing video games with the real Commodore or compatible joysticks, or doing cross-development where you develop your code in VICE and then you can test it with real hardware controlling it, or vice-versa – test your C64 developed code in the emulator with single-step debugging and the real interface.

The only gotcha for new users is that VICE must be set to use a “positional” key map instead of the default “symbolic”. This setting is under Settings → Keyboard → Keyboard Mapping. It will still work just fine, but VICE will interpret each key incorrectly and show the wrong character.

Oh, and because this shows up as a standard USB HID device, you can use it with the VIC-20/C128 emulators, and you can also plug it in to a C128 where it will boot it into C64 mode. It also works with the Ultimate 64. It should even be usable as a regular keyboard input device, so you can write your next novel on your C64 keyboard. Unfortunately, the VIC-20 uses a totally different set of expansion headers, so while you can use your C64/C128 to control the VIC emulator, you can’t use the VIC-20 hardware with this kit. Perhaps a sibling project might show up down the line?

It’s a simple, easy to solder through-hole kit. The chips come with sockets, and so future firmware updates will be very easy to install for end-users. The entire project is fully open-source (woo!) and is available under CERN OHL-P v2 at the GitHub repository.

source https://blog.tindie.com/2026/07/give-c64-emulation-a-realistic-touch/

Lua-based Microcontroller Platform

There are lots of alternatives for programming microcontrollers other than C/C++. Python has become very popular, with MicroPython and CircuitPython exploding in popularity in recent years. But Lua has always had a presence in the embedded field. It’s an exceptionally flexible scripting language, giving users the power to essentially make any data structure they want from Lua tables. The ELM11 Microcontroller Board from Brisbane Silicon is a powerful, multi-core chip specifically designed to quickly interpret Lua.

I had the opportunity to play with a couple of these boards over the last months, and they are extremely cool! Not only has the documentation been continually improving, but Brisbane Silicon recently launched a new IDE specifically for developing for this board. This simplifies a lot of the process, especially in creating “hardware overlays” which are custom implementations of whatever hardware interfaces you need for your project — from the basic serial buses like SPI to whatever you can dream up, if you’re willing to roll up your sleeves and design your own.

It’s easy to get up and running quickly. When plugged in to a computer, you get a serial port that allows direct access to the Lua REPL running internally. You can play around with importing different modules and prototyping basic code. When you’re ready to push a project to the board, you can use either their new Arvore IDE or the original command line tools. Their website has a ton of information about using and programming this chip.

The ELM11 is the brainchild of two guys from, you guessed it, Brisbane, Australia. Craig Haywood designed the software side, and I asked him why they chose Lua for this project. His response was “Lua is quite simple to use, and it’s great for low-resource environments like embedded platforms. It’s also easily extensible; users can rapidly and easily add their own modules and expand functionality as they see fit. And because Lua is such a lightweight language, implementing a lightning-fast interpreter is not only feasible but easier to do than with other scripting languages.” I couldn’t agree more – I’ve always had a soft spot for Lua, and if you are curious about the board, make sure to check out the product page!

source https://blog.tindie.com/2026/07/lua-based-microcontroller-platform/

Show off your creations with Tindie at Teardown 2026!

Teardown 2026 is happening in Portland, Oregon this July 24-26 at the Jupiter hotel. It’s going to be a fun event, with talks, workshops, breakout sessions, and tons of cool hacks. In collaboration with Open Boards Guide we will be there and we want to show off your cool Tindie projects!

So, if you’d like to showcase your product there, all you need to do is post your project in this thread over on Tindie’s forums, and propose anything you’d like to have featured. We’ll pick the coolest stuff, and if selected, you ship us a sample of your project. It’ll be featured in our display at Teardown, with info about your board and a QR code directing people to your product page. If you have any issues with this process, email David at tindie@ishotjr.net for help!

This should get lots of eyes on your project! Once the event is over, we’ll ship everything you sent us back to you. If you happen to live in the area and can attend the event — even better! Grab tickets, bring your project with you, and you can be present to promote your gadget and answer any questions that users may have. For those who can’t be present, try and give us the most frequently asked questions and info about your project so we can represent you and show people what your product does!

If you want to attend Teardown, tickets are still available here and it’s going to be a great event. Check out the full schedule for info on the speakers, topics, and workshops that will be happening.

source https://blog.tindie.com/2026/07/show-off-your-creations-with-tindie-at-teardown-2026/

Important Update: Tindie’s New Payment and Seller Payout System Is Now Live

Hi everyone,

Today, we are officially launching Tindie’s new payment and seller payout system. This is an important change for the platform, and we want to explain why it was necessary, what is now available, and what sellers should expect during the transition.

First, thank you for your patience over the past few months. We know the payout situation has caused frustration, delays, and additional work for many sellers. We are sincerely sorry for the inconvenience.

Why we had to change the payout system

For more than two months, we worked continuously with PayPal to restore the automated Payouts function previously used by Tindie. We submitted extensive company, identity, transaction, order, fulfillment, and shipping documentation, and made more than one hundred calls to different PayPal support teams. At several points, we were told that Payouts had been approved or enabled. However, actual seller payouts through the API continued to be blocked by PayPal’s internal risk systems.

The main difficulty is that our PayPal account is new, while Tindie is an established marketplace that immediately began processing a large volume of transactions. Tindie is also not a traditional retailer: products are stocked and shipped directly by independent sellers, so the platform itself cannot provide fulfillment documents for every order. After repeated reviews, temporary approvals, new restrictions, and additional document requests, it became clear that PayPal was unlikely to provide us with a stable automated payout solution in the near term.

We could not ask sellers to wait indefinitely, so we had no practical choice but to introduce a new system, which we’re announcing today! Some sellers may have already noticed the changes on site. You now have two options for payouts.

What is now available

Stripe Connect

Stripe Connect is now Tindie’s primary automated payout method. Sellers in supported countries can connect or create a Stripe account, complete the required verification, and receive eligible payouts automatically. Availability, verification requirements, and payout timing may vary by country. Once you have established a Stripe account, it should be smooth sailing moving forward for payouts.

Wise

Wise is available as a manual payout option for sellers who cannot use Stripe Connect or prefer Wise. Wise is an online payment processor that we have used internally and had great experiences with. It’s easy to set up an account, and you can easily convert between different currencies and then either use a Wise card directly or transfer funds to other bank accounts. Wise payout requests will have to be reviewed and processed manually by the Tindie team, though, so it may not be as fast as Stripe Connect. Sellers may be asked to provide a Wise payment link or other account details in order to have funds disbursed.

Updated payment and marketplace information

Along with the new payment system, we have also updated several important pages on Tindie:

• Buying on Tindie
• Selling on Tindie
• The Tindie Guarantee
• Terms of Service

These pages provide clearer information about payments, seller payouts, fees, shipping responsibilities, refunds, buyer protection, and the roles of Tindie, Stripe, Wise, buyers, and sellers.

We encourage everyone to review the updated information.

A clearer and more sustainable payment structure

This update is not only a change of providers. It also allows us to make Tindie’s payment and payout structure clearer and more sustainable.

The real cost of international marketplace transactions can vary significantly depending on the buyer’s country, payment method, currency conversion, refunds, and payout method.

Under the previous system, these costs were not always reflected clearly and were sometimes absorbed by the platform. That approach was difficult to sustain and limited our ability to invest in support, security, infrastructure, maintenance, and new features.

The updated structure is designed to provide clearer fee information, more reliable seller payouts, and a healthier foundation for Tindie’s long-term growth.

Please expect a transition period

During the past two months, our technical team has also been:

• stabilizing Tindie after migration to our infrastructure;
• repairing problems in an older system with many third-party plugins;
• responding to large volumes of malicious traffic and attacks;
• implementing and testing the new payment system;
• manually processing seller payouts during the transition.

We have done everything reasonably possible to make the new system stable before launch. Even so, some sellers may encounter verification requests, country-specific limitations, payout delays, or interface issues during the first days and weeks. If you experience a problem, please contact Tindie Support and include your store name, country, payout method, and a screenshot or clear description of the issue. Our team will work with you to find the most suitable available solution.

We understand that changing payment methods may be inconvenient, especially for sellers who have used PayPal for many years. We are sorry for the additional steps, and we appreciate your understanding and support. Unfortunately, we are left with no choice after trying to find a solution with PayPal.

Looking ahead

Launching the new payment system is an important step, but it is only the beginning. Our immediate priorities are stable platform operation, reliable seller payouts, clearer fee information, faster support, stronger security, and better tools for sellers and buyers. Once these foundations are stable, we will continue adding new features and connecting Tindie with more industry resources.

Our goal is to make Tindie a stronger home for makers around the world — a place where creative hardware ideas can become real products more efficiently and reach the people who need them.

Thank you for your patience, feedback, and continued support as we take this step forward together.

— The Tindie Team

source https://blog.tindie.com/2026/07/important-update-tindies-new-payment-and-seller-payout-system-is-now-live/

May 12th Update

Quick updates for everyone on the disbursements. We got all set up with PayPal after transferring all the account info over, and we were able to send some payments out but apparently we tripped some kind of safety mechanism with PayPal — we tried to send too much, too quickly as a new account and so we’ve had to go through a long, drawn-out manual review process. We are able to manually disburse funds now but automatic disbursements will take a while. We want to be careful to not get locked out again!

The staff running the support@tindie.com email assure me that they have been reading and responding to all emails they receive, and constantly check Spam folders etc. just in case. To be safe we’re investigating whether there is some other issue as some people have reported not getting a response. If you don’t get a response, maybe try from a different email address, or try again after 12-24 hours. If you are sure your emails are not getting answered, try from another address or contact me on social media and I can pass your email address along for investigation.

All sellers will be disbursed — we are disbursing sellers manually at the moment which requires some manual labour in each case (checking all the info given to us previously and checking for sales since then) and this will happen more quickly as we “gain trust” in the PayPal system. We’re working as fast as possible to resolve this!

source https://blog.tindie.com/2026/05/may-12th-update/