I wish it could write integration tests. That requires an understanding of the testing scaffold and the fake data it produces. LLMs aren't ready for that yet, but someday it will be magical.
Sounds like you've been unfortunate and have spent all your time with financially driven people. A bunch of that might from from working in advertising. I can assure you that there are founders and companies that are ethical and moral, though they may be rare. My point being that you might want to focus on changing the type of people you work with as much as the topic and market segment. Certainly you will find a higher percentage of people motivated to do good for the world in the nonprofit sector, for example.
One of the pernicious aspects of using AI is the feeling it gives you that you have done all work without any of the effort. But the time of takes to digest and summarize an article as a human requires a deep injestion of the concepts. The process is what helps you understand. The AI summary might be better, and didn't take any time, but you don't understand any of it since you didn't do the work. It's similar to the effect of telling people you will do a task, which gives your brain the same endorphins as actually doing the task, resulting in a lower chance that the task ever gets done.
We've had the capability to feed the entire planet for decades now, and yet, a large portion is underfed. Even if Sam's wildest dreams come true, this sounds like another "divide the world into haves and have-nots". One way to justify this unethical view is to consider that his audience was a set of target customers, rather than "people in the world". IOW, this was just fancy marketing fluff.
Okay, so, GPUs will go away to be replaced by AI, which... runs on GPUs? Even if the target application changes from games to AI, we still need widely parallel processors (aka GPUs) and serial processors (aka CPUs) if we want to keep growing performance.
You can't enforce silly laws. Who would enforce this ban? Teachers? That would be a school regulation, not a state law. Or do you want police walking around school arresting kids with phones? Do you enforce all of your guidance for your kids using legislation? A real solution to the distractions and harm that phones cause when used at school is harder, of course.
Check out the book Traction, to get a bunch of free marketing channels and a strategy. My favorite is to go after the bloggers that are read by your target customer community.
The reasons GPUs are useful is that they allow us to stay closer Moore's curve. Power is the problem, not AI algorithms. AI algorithms are a tensor processing best case right now. It took us decades to develop widely parallel algorithms for graphics rendering. AI will develop new algorithms. And algorithms will be optimized to be more sparse, which will make them less of an ideal case for the widely parallel hardware. But we will likely need both serial and widely parallel cores to handle any future algorithm.
Glad to hear money isn't your focus. Your comment "...then we realized it was also monetizable" was the reason for my interpretation. Its also a very common rational. I don't know what "nerd-sniped" means, so...
Good luck with the VCs. I hope you all stay friends through the challenging process.
You said (implied?) that your reason for starting a company was that you were waiting for somebody (MS) to build your favorite tech, and you realized it was monetizable. Finding a gap is a great start. But, if money is your goal, it is far easier to make money working at a company than starting one. Existing companies are great places to learn about technology, business, and the issues that should really drive your desire to start something yourself.
I recommend getting a job at NVIDIA. They care deeply about SW. It is a great place to learn about HW and the supporting SW. There is much to learn. Maybe you will learn why you are unimpressed with their SW offerings. For me, the hard part was the long lead time (8+ years) from design to customers using the product. One of the things that always amazed me about NVIDIA was that so many of the senior architects, who have no financial need to keep working (true for more than a decade), are still working there because they need the company to do what they love.
Why start a company to make this product? Why not go work at one of the existing chip manufacturers? You'd learn a ton, get to design and work on HW and/or SW, and not have to do the million other things required to start a company.
And the auto and Defense markets are so easy to enter! /s
Both of these markets have long lead times, tight HW build times, and move incredibly slowly. They are not the kind of markets that like using stuff from new companies with no history. Again, I'm no expert, but I'd say you need to be concentrating on sales and market research now.
I think the only way this could work is if you had the backing of one of the major LLM providers who decided that your ideas are worth doing a PoC. That way you actually have a client on board before you spend all the money. I know you guys probably like the designing of the HW and SW, and maybe the implementation of both, but really, what you need now is to do sales.
I cannot see any way of building HW profitably for the Jetson market. You are really competing with Raspberry PI, not Jetson, IMO. I mean, I'm no expert, but I would suggest doing a deep dive on your business plan if you intend to target the small hardware world rather than spending any time designing HW or SW. Then reduce your estimate by at least half since doing anything in that embedded/edge world has many more technical issues.
In my experience, trying to switch VFX companies from CPU-based rendering to GPU-based rendering 10+ years ago, a 2-5x performance improvement wasn't enough. We even provided a compatible renderer that accepted Renderman files and generated matching images. Given the rate of improvement of standard hardware (CPUs in our case, and GPU-based inference in yours), a 2-5x improvement will only last a few years, and the effort to get there is large (even larger in your case). Plus, I doubt you'll be able to get your HW everywhere (i.e. mobile) where inference is important, which means they'll need to support their existing and your new SW stack. The other issue is entirely non-technical, and may be an even bigger blocker -- switching the infrastructure of a major LLM provider to a new upstart is just plain risky. If you do a fantastic job, though, you should get aquahired, probably with a small individual bonus, not enough to pay off your investors.
Absolutely. Startups should not hire junior engineers. They don't have time to teach people until they are a functioning profitable business.
However, I would not encourage learning at an agency or consulting service. The best place to learn by far is at a large corporate company, where you have the time to learn good programming practices, and people to teach you how to do it well. The problem with consulting is that you're always working on a short-term issue, and never learn the value of writing good code for long-term support.
Nothing will really change. Even if they are found to be a monopoly. The appeals process and the commensurate change in the world during those years will allow Google to work around any proposed remediation before it is enacted.
You can scale revenue without growing the number of employees. That's precisely why product-based businesses, like SaaS and software in general, are famous. Service-based businesses require scaling employees in order to scale revenue. In fact, if you keep employee growth in check, each employee makes even more money. Further, if you split profits, rather than forcing your employees to wait for the mythical IPO, they can use some of that money much earlier than if you grow employees, take on more investment, generate a huge overhang, give your employees low "startup" salaries, and then get aquahired with common equity getting zero return. In practice, it is far easier to get big payoffs for your employees if you keep the employee count small and aim for an early exit.
The only way to win is not to scale. Most of the problems discussed by PG and in this post revolve around scaling company culture, and the associated management style, as the company grows the number of employees. Unfortunately, my feeling is that there really is no great solution to communication problems after a company scales to 20, or 50, or 200 (choose your favorite) people. Modern tools are helping more and more (GDocs, calendars, meeting transcription, ...), but after you get past 15-25 people, you go from everyone knowing everything to something less good. There's no reason that companies "need to scale". You can make millions of dollars without scaling up, and you can avoid many of the problems that kill companies at the same time -- investment overhang, cultural changes, management challenges, and the need to expand your customer base.
The best part are the follow-up emails of people trying to cover their ass and pass the buck within the giant corporation. I'm sure all those emails and the time it took to read and respond to all of them, and all the meetings about them that we don't see, really helped the world. /s
That second paragraph, "hungrier and greedier", is the problem with the entire capitalistic system. If only people stayed small, they'd be happier, have more time, and there would be a wider more diverse set of products available for the public. More importantly, staying small avoids many of the problems startups often hit. It's much easier to handle the culture and internal communications, you don't need investment, and you can focus on making a fantastic product for your much more cohesive set of customers.
Also ironically, kinda like English, which is ugly, and has even more ways of saying anything. Language and writers, have done far more damage to the world historically than coding or programmers. Also many great and wonderful things, just like coding. Outrage is such a wonderful emotion.