Nov 28, 2022
What does it take to solve an intimidating problem that many feel
is unsolvable?
Andrew Feldman, Co-founder and CEO of Cerebras Systems, can tell you, because he and his team engineered an unprecedented technological breakthrough. They set out to build a new class of computer system to accelerate Artificial Intelligence work. In the end, they built the fastest AI accelerator, based on the largest processor in the industry.
Tune in to hear his story, his thoughts on building and selling companies, and his career advice for aspiring founders.
Enjoy this episode.
Main Takeaways:
- Pioneering Solutions to Big Problems: Andrew explains his love of
tackling big problems where there “isn’t a safety net” and his love
of “fearless engineering.” He shares his experience searching for a
solution that many thought couldn’t be found.
- Artificial Intelligence: Andrew discusses his thoughts on AI, how
it could be used for extraordinary good and how it will permeate
every facet of our lives moving forward.
- Building and Selling Companies: Andrew discusses his experience
building, leading and selling companies, diving into his decision
making process for when to sell.
- Advice for Aspiring Entrepreneurs: Andrew provides advice for
aspiring entrepreneurs and also dives into the mistakes that he has
made along the way. He stresses the need to focus on the customer,
to build trust within your team and to build the right professional
network.
Key Quotes:
In my experience, these aren't sitting on a park bench with an idea
arriving like a child from Zeus's head fully formed, right? That's
not the way they come. You articulate a problem….What choices are
available to you to solve this?
You have to decide in your career what sort of problems you're going to attack. And you have to decide if you're more afraid of failing in pursuit of a really interesting big problem or succeeding at a mediocre problem.
I think like every technology, [AI] has the opportunity for tremendous good and tremendous evil, both. I'd say the same for nuclear power. I'd say the same for, you know, any number of monstrous technologies. It is that their very power can be used for good or for bad. And I think that the technologies in AI can be used for evil and the exact same technology can be used for such good, it's extraordinary. And so the challenge is on us to manage it.
I think it's a tremendous mistake to build a company to sell it. I think it's a tremendous mistake to have a religious view that you have to go public. You are using other people's money in what we do. You are building a company in partnership with people who are lending you part of their career, and you're the steward of that.
I think one of the things young people should think about is they see resumes and they see LinkedIn links and it's success, lots of bullets, another success, lots of bullets. I think you can just ignore all that because nobody puts up their failures. You know, bad idea, six months wasted on a bad idea, millions of dollars destroyed because of arrogance, right? Nobody puts that on their LinkedIn. And so you get this, sort of like the Instagram version of a career, perfect angles, perfect lighting filters done properly. But that's not really the way careers went.