Shipping 100 Hardware Units in Under Eight Weeks
Posted4 months agoActive3 months ago
farhanhossain.substack.comTechstoryHigh profile
controversialmixed
Debate
70/100
Hardware DevelopmentVoice AssistantsDriver Safety
Key topics
Hardware Development
Voice Assistants
Driver Safety
The author shares their experience of shipping 100 hardware units of a voice assistant device in under 8 weeks, sparking discussion on the tech's potential, safety concerns, and the product's viability in the market.
Snapshot generated from the HN discussion
Discussion Activity
Very active discussionFirst comment
N/A
Peak period
22
18-24h
Avg / period
8.5
Comment distribution94 data points
Loading chart...
Based on 94 loaded comments
Key moments
- 01Story posted
Sep 18, 2025 at 4:11 PM EDT
4 months ago
Step 01 - 02First comment
Sep 18, 2025 at 4:11 PM EDT
0s after posting
Step 02 - 03Peak activity
22 comments in 18-24h
Hottest window of the conversation
Step 03 - 04Latest activity
Sep 22, 2025 at 1:35 PM EDT
3 months ago
Step 04
Generating AI Summary...
Analyzing up to 500 comments to identify key contributors and discussion patterns
ID: 45294390Type: storyLast synced: 11/20/2025, 5:51:32 PM
Want the full context?
Jump to the original sources
Read the primary article or dive into the live Hacker News thread when you're ready.
About me: I’m a robotics and product design engineer focused on building thoughtful tools for the world. I hold dozens of patents in hardware and manufacturing, and I care deeply about how things are made and who they’re made for.
For over a decade, I’ve worked across robotics, wearables, and consumer electronics. As one of the first engineers on the Apple Vision Pro, I took it from concept to mass production.
I'm also in the wearables space, though neurotech/sleeptech.
I'm assuming you did 3d printed enclosures, so really board design and was the longest process.
What I think is really clever about your design is passthrough USB-C and then not needing your own battery. So essentially you've got a micro, probably with it's own memory?
So elegant.
Others are saying you must have had your Taiwan contacts beforehand, but even without that, two weeks for board manufacturing isn't unrealistic I'd think, even for a noob, and lucky for you the board design should have been pretty simple.
Can I ask what your experience going through YC as a consumer hardware founder was like?
If you're curious about what we're building, we're enhancing the restorative function of sleep, without altering sleep time. Check out https://affectablesleep.com
I think for consumer, if you can really simplify the product and solve the absolute basic version, the costs should be low enough to validate the idea. YC will value your skills to create this simple version, and that you are able to actually execute and create something that could be real.
The missing link was really showing I could take a prototype and mass produce it (even at a small scale). That was what this whole exercise was about.
One additional note that comes to mind, building really great partnerships is essential for hardware to work.
My previous start-up was acquired
Our technology is in use in clinical trials. We've begun pre-sales, and are in selection of contract manufacturer.
I only decided to apply for the next round last night. Would it be ok if I ask to get your opinion on a few of the application questions? When we get to that point? Probably not for a few weeks.
We're fortunate that our hardware engineer and industrial engineer both have extensive experience in manufacturing.
But it seems like a terrible hack - using the phone through the limited interface designed for humans instead of actually using the proper APIs. Not to say that any effort from Apple to improve Siri will just render this product obsolete instantly.
Could you identify any specific decisions/guidelines early in your career that led to your success today? I'm currently studying engineering (will likely specialize electronic/embedded) and am always looking for ways to improve.
It eventually needs to do something the phone software cant do itself. For example more powerful AI chip than the phone has.
We live in a physical world, and some of us should build things for it.
Joyce flew from Taiwan to make sure I had them in my hands, folded boxes with me in the office, and just as she got over her jet lag and went back to Taiwan.
The tech seems neat and all but please stop multitasking while driving, encouraging others to multitask while driving, and building products specifically designed to encourage multitasking while driving.
If you want to work while you're in transit: take public transit.
In 2025 we can’t spend 10 minutes without doing something else while traveling at speeds that would make a sailor blush.
Try it out - watching this is sooo soothing and relaxing. :-)
Building tech is usually clearer than finding a clear use case for it. As we find ways to mature the tech to be aligned with the ultimate vision we have, we will test various problems the immature tech can solve.
With that being said, if you have any ideas where this could really help people (for instance people with motor disabilities), please share them. We would like to serve people and build with humility.
Lots of folks in safety critical situations rely on multitasking and voice commands: law enforcement, firefighters, pilots, heavy equipment operators, armed forces, etc. Many of them are in situations in which not multitasking isn't an option and receive special training to minimize risks. That being said now you're entering heavily regulated industries where the stakes couldn't be higher... not exactly an easy place for LLMs and startups to play.
On the other end of the spectrum there is a tried and true industry for tech innovations where the stakes couldn't be lower: adult entertainment. There's millions of adults wishing they could operate screens without needing a hand free. Might not be as glamorous as helping firefighters and people with motor disabilities, but we all need to make a living.
Best wishes!
That being said, my bike ride to work is a quiet 30 mins. This would be insanely useful for that. I always get to work with a stack of messages piled up, both ways.
People do not multitask well. The median driver can barely keep a car on the road in perfect weather on a good day. Please stop helping distract them!
Luckily, most states lave laws about this, and drivers will get fined. Sadly, manufacturers rarely get fined. The people who made the original "hands on wheel" defeat device for Teslas got away with it, IIRC
I remember an experiment about driving there the subjects had to go to a simulator and drive the same simulated route for some time, until they know it and it got boring. So, after a few days, suddenly a deer/pothole/kid/car/whatever appeared.
In the bay though. Holy crap driving here was a wake up call for me. You cannot drive distracted here. I've driven in every major US city. The bay is by far the worst. Yes even then NYC, ATL and Miami. Chicago or Houston is a probably 2nd worst.
At the same time my partner often gets mad at me for refusing to do anything but focus when I drive. So I agree in general but I'm not everyone. This device seems like it will help people avoid incidents.
Quickly, before you kill someone
They're saying you always need to pay attention.
I'm not saying its right, I'm straight up militant in my driving. I'm saying its part of human nature. Anytime you fight that you are going to lose. So a product that reduces the distractions even a bit can be a net good thing.
I hate to be this guy but I've lost too many people for small stupid mistakes. The equivalent of this is getting killed when you accidentally bump into someone in the supermarket. This is just not fair.
Never take a chance, stay safe, life is precious and fragile.
I concur with the other comments, this is a "yes dear" moment that you'll appreciate once you're older.
Or worse, someone else.
If someone kills themselves because they are driving distracted and they chose to distract themselves, that's just evolution IMO. But if anyone else is hurt, then it is a tragedy.
I dont drive distracted but i can understand why others do. I like anything that helps them focus more.
On the one hand, you might get a few people who would be handling their phone to keep their eyes on the road more.
On the other hand, it might facilitate phone use in a way that people might use their phone now when they wouldn't have before (because handling it felt too distracting but this doesn't). I think this is a tangible risk.
There have been papers published that point out that hands free use isn't anywhere near safe.
Side note: not sure why the downvotes or what the other somewhat strident replies are about. I don't quite agree with you but it's not downvote worthy.
HANG UP AND DRIVE!
Don't talk, don't try to do anything other maybe listening to some music while you're driving.
It's not very effective!
How are you going to get drunk people out of cars and onto trains? We've already made it really fucking shit for them. What's your next idea after trains? Prohibition?
I'm contributing to a similar open source coding tool [1] and I see the same skewed reaction: voice control of "whatever" while driving gets 5-10x the clicks of any other demo.
There's a logical reason so many people think of voice control while driving. It's not because they're reckless.
It reflects the hierarchy of needs. People with long commutes (often younger, lower-paid engineers living further out) spend 2+ hours driving daily.
This is their biggest time sink, so of course they think about making it productive. When you're living far out for cheap housing and hear "coding while driving", its easy to think: finally, a way to get ahead without choosing between career growth and seeing my family.
Again, I think its just an off-the-cuff reaction, not actually what people will do. Just like people try your app and tell you its amazing but then never pay. Doing stuff while driving just sounds nice until you know... you think about it for 3 seconds and yeah, its bad idea.
[1] https://github.com/slopus/happy
[1] https://www.youtube.com/watch?v=-zRN7XLCRhc&t=1981s
To me the point of voice control is walking. I'm thinking of Einstein, Darwin, Thoreau. They believed that physical activity helped stimulate his mind and spark creative thought.
You walk, you think, and occasionally you say something.
But in practice people have been quick to jump to conclusions that surprise me. Such as "the point is to try to kill people".
So I'm merely working backwards to figure out why people want to jump in a car first thing.
And the charitable explanation is the people are just not thinking to carefully and are just excited by seeing a computer respond to voice with a low enough error rate.
But the blue founders here don't fall into the first category, and they made a 2 minute ad advertising their product that doesn't fall into the second category either. Charity only stretches so far.
I'm not a startup investor, but I'm quite serious when I say that ad would give me serious doubts about doing business with this founder for purely practical reasons (on top of other ethical ones). That doesn't mean the product is bad, but a business is more than a product.
---
HN definitely does have a dog piling problem where one issue is picked out by the community and discussed ad nauseam to exclusion of all else. To be honest between the magnitude of this issue and the fairly significant discussion downthread about other topics I'm not particularly feeling that it is a problem in this thread though.
It really isn't. Driving while distracted by a phone conversation is comparable in safety to driving while heavily intoxicated or while barely awake. This is not a hypothetical. There is actual experimental research behind it and the evidence is consistent and overwhelming. There is no justification and no defence. It's not a debate. It kills people.
Everyone is right in saying that those drivers should take public transport instead if they want to work or call or play a game while they are on the move. And yes - maybe in time there will be self-driving options that will make the whole issue obsolete. But right now anyone driving but not paying full attention is a danger to themselves and others and this is not something we should condone or attempt to justify. It kills people.
In some ways, if this instigates them implement it in their OS, we are doing our job, and then we can pivot and keep working on ideas that we hope will serve others.
One obvious dumb question is is there a video input for the device? I mean how does it know what screen it's on, where to find the buttons to click. It seems like the next level would be to add a video capture device too
For example, if you're old enough, do you remember those devices that were all about getting your phone/ipod with a 3.5mm jack to interface with your car audio system? There were 3.5mm -> cassette adapters, and bluetooth -> 3.5mm dongles abound before bluetooth became commonplace in cars and phones.
If all you are trying to do is finding the right podcast episode, it might be quite a bit better than fiddling with a screen.
But... cognitive load is a thing.
https://road-safety.transport.ec.europa.eu/system/files/2021...
The reason conversations can be held safely between the driver and with (some! people are different) other occupants in the same car, is that all occupants can asses what is happening and react to microcues from the driver in realtime. (Phone calls add many milliseconds in reaction time, plus the other person is not situationally aware.)
I do think an AI could help here though. This app isn't quite that. The AI would have to be very fast and responsive, notice with cameras and microphones when the driver needs a microbreak in the conversation, seamlessly continue when appropriate, perhaps be plugged in to the car radar and so on. Really be a polite and attentive passenger.
I was confused by how CarPlay was working, as this requires the vehicle to be a host and the phone a device, but I'll assume it means wireless CarPlay (vs USB 2.0 CarPlay).
For the display, I wonder if the device is mimicking the assistive touch inputs via external USB PID (per the note) and using a mouse/trackpad HID can take a screenshot via assistive framework - I guess the app could see this arriving, process it and then delete? Also figure out what is being shown? would explain why the video is sped up a little, but YIKES. Also good luck stopping the app being killed in the background.
All direct USB streams require MFI auth, which might work for exactly 100 devices and could be an alternative, but requires Apple cert for bulk quantities.
I think ReplayKit can only work within an app and can't see the general output stream.
Cool use of tech and curious to see how it actually works :)
But, this is one step closer to my dream of being able to do my software engineer job while on walk through a park. Chatting async with an AI coding agent. Sitting on bench to review the changes. Repeat.
Y’all made a lot of smart choices to keep your timetable. Not least of which was finding - and listening to - a good CM partner.
So if this guess is correct then if android allowed screengrabbing and sending mouse events, the hardware would not be necessary.
BUT, this will soon be a dead thing once these agents land in devices natively (like in <1year or so). So use this expertise you gained at creating a hardware and redirect it elsewhere instead of doubling down.