How to Learn Hardware

Last year’s post “You Should be Working on Hardware” was mostly targeted at people with established careers and financial security, but I was surprised to see it enjoyed much wider readership.

Disclaimer: My own path to hardware has been personal, contingent, and path dependent. Do not feel obliged to validate my mistakes by repeating them. Also, hardware businesses are objectively much more capital intensive, risky, and challenging to execute. If you’re optimizing over expected returns in a finite lifetime, the best strategy is to pick the highest return thing with a decent chance of success for you. Starting some insanely high risk hardware company is probably not that. Elon Musk thought he would probably fail at both Tesla and SpaceX but that it was worth trying anyway – but he would have a soft enough landing even if he failed.

Anyway, lots of people have asked me how to go about learning hardware. This places me in a predicament, because I don’t know much about hardware – I started a hardware company in part to learn as much as I could as quickly as I could – with some success. I also learned more than I bargained for about labor law compliance and facilities management. Well, it’s all part of the puzzle.

So, in no particular order, some ideas I have found helpful.

You need to accept that learning new things as an adult who is expert in other areas is not always fun.

YouTuber AvE has a couple of nice videos explaining the learning curve. The key insight is that the fun part of the learning curve, where you are gaining skills quickly, comes after hours of painful mind numbing frustrating struggle. Hundreds of hours. I learned to ride a unicycle in about 5 hours – anyone can. You will fall 100 times – everyone does. You will probably bleed – that is the cost of gaining the lifetime ability to zoom on one wheel.

Grant Sanderson’s commencement speech has some similar insights, in particular: “Action precedes motivation.”

In my mispent youth I burned off some excess testosterone by hitchhiking around the world a few times, over two or three trips, each months long, on other continents, with strange cultures and unfamiliar languages, before cell phones. On each of these trips, the hardest part by far was closing my parent’s front door behind me and stepping beyond the comfort zone. After that, it was one foot in front of the other until I returned – but I never really did. My childhood self was transmuted into something different out on the steppe.

Sanderson has some other insights (paraphrasing slightly): “When you’re a student the fundamental goal is to grow, to improve, to become better. In the real world, success hinges on how effectively you are able to add value to others.” “Your dream has to be externally focused – service, rather than self serving. People who succeed in their early jobs make life easier for the people around them.”

This is an important insight. Skills, leadership, self-sufficiency. It’s fundamentally about service to others. This insight ranges from the local and minute to the cosmic. Every successful business is a corporate entity that survives by adding value for other people.

Write a blog. Document your progress. Why? It creates a proof of thought, a proof of work, and makes it easy for other like-minded people to find you, even years after the fact. You can make videos and other artifacts too.

Where do you get your dopamine? This would appear to contradict the earlier points about finding the toughness to do hard things. But it doesn’t really! It is always easier to learn things you enjoy doing. The art lies in finding ways to enjoy the things that are necessary, even if you have to approach it abstractly as Type 2 fun. And finding ways to avoid enjoying to excess things that are counterproductive to your mission in life – usually addiction in various forms.

Judgment is a skill and it can be learned. Hardware development in a start up setting is insanely resource constrained. It helps to avoid making lots of expensive mistakes. Here are some insights I’ve used to improve reasoning. Problem solving in physics – the algorithm. Also, a note on inference.

Safety. You should be working on hardware because one day you will die, but you don’t want to die, or be maimed, because you were working on hardware. More than one hydrogen entrepreneur has blown themselves up in their garage. Plenty of people electrocute themselves every day. Basic PPE and safety training will reduce the odds of an unplanned trip to the ER by factors of thousands. Use your recently honed judgment skills and protect your money makers!

For focused learning of particular verticals, there are two complimentary approaches that can be taken.

Skill-focused development – oriented around a particular topic or tool set. For example, wood working, using a lathe, learning CAD. Back in the day, you’d need to find a mentor or a coach. These days, YouTube and various LLMs will make it far easier to bootstrap your knowledge.

Outcome-focused development – oriented around a personal goal. For example, what tools and skills do I need to change the oil on my car? Fix a plumbing issue in my house? Safely climb a tree? Launch a rocket? Build a plane? If your hardware vision revolves around a specific vision for the future that will otherwise not occur if you don’t do it, that could be a good jumping off point for learning stuff.

“Should I go back to school?” In general, no, unless you need a particular certification on your journey. It is possible to learn almost anything much faster and cheaper in a focused way than in college. On the other hand, college has a much higher completion rate than Udacity and similar, probably because it’s much easier to endure hundreds of hours of struggling with material and feeling like an idiot if you have friends who are there with you, commiserating and creating a social cost for slacking off. There are some hardware skills (eg electrician, plumber) that are fairly easy to learn through a trade school, and others where there is no substitute for getting dirt under your nails. In general, with hardware stuff, you need to get your hands dirty early and often. You can’t learn it from a book.

“I want to start a hardware company, do I need to know that much about hardware since a) I can hire engineers and b) as a founder I’ll spend most of my time in meetings anyway?” It is certainly possible to be a non-technical (co)founder of a hardware company. Richard Branson founded Virgin Galactic, which has successfully launched unique orbital class rockets! But his lack of hardware expertise, instinct, intuition, etc probably did not help. Elon Musk attributes some of his success to leading (initially against his will) the technical as well as business parts of both Tesla and SpaceX. It allows much faster resource allocation decisions, and most companies die not because they make bad decisions, but because they stop making decisions at all. It’s also much easier to hire good engineers if you can speak their language.

“Can I learn hardware on the job?” Maybe? If you’re a software engineer at a big company that also does hardware, you’ll probably be expected to focus your efforts on stuff where you have high leverage productivity, but some places will allow you to cross train. If your company does hardware development, getting lunch with the workshop crew will help you meet people who may be able to give you some advice – if you respect their cultural and social norms! But you can’t learn hardware over coffee, at least not very quickly. The only companies that can afford to tolerate generalists screwing around in the lab are desperate startups who don’t have the resources to hire a large team of specialists. But for most hardware beginners, a $50 kit of parts from Amazon and a desk somewhere is easily enough for months of enjoyable frustration.

I will probably update this post and add to it, particularly in the coming weeks, as more people contribute insights and questions. My intent is not that this is comprehensive. The single bit insight is to convey a bias towards physical action with hardware.