Groq kept delivering so their valuation has effectively gone up.

A year ago it wasn't clear if they'd stay competitive but it seems they are.

>> if groq and cerebras combined

There isn't to be shared between the two techs, Groq's hardware is a like a railgun that installs all the weights into the optimal location before firing off an inference. Cerebras computer engineering more convention requiring the same data movement that GPUs struggle with optimizing.

Suspect Groq is complementary/superior to nvidia's GPUs, while it is unclear what Cerebras brings other then maybe some deals with TSMC.

They are both SRAM based solutions currently with the same benefits and pitfalls.

Indeed. Seems more like a hit piece to me. Amazon is a company the Guardian often likes to pick on.

Yeah we generally don’t give those permissions to agent based coding tools.

Typically running something like git would be an opt in permission.

The next hype-cycle might not relate to software. I'm thinking of the (smaller, shorter) phase where "nanotechnology" was getting slapped onto everything including laundry detergent.

mRNA was set to be huge but US voters apparently didnt want it.

Defense will increasingly become a national priority over the next few decades. Pax Americana is teetering.

That and consumer robotics. The latter will explode if (big if) RL and llm reasoning get combined into something solid. Lots and lots of smart people are working on it already of course, we are seeing great improvements but nothing really usable. i think we will finally get to a real hype stage in maybe 3-4 years

That’s the $100M question. Pick the answer correctly and you’ll get record breaking compensation levels.

Probably defense and robotics.

The US mix of government and enterprise. You want Capital you need to be MAGA

"Donald Trump and Silicon Valley's Billionaire Elegy" - https://www.wired.com/story/donald-trump-and-silicon-valleys...

"Secret White House spreadsheet ranks US companies based on loyalty to Trump" - https://www.telegraph.co.uk/business/2025/08/15/secret-white...

Smartest people will be working against that. You're thinking of opportunistic people with myopia.

Pretty much the entire tech industry has bent the knee by now - they even gifted the new ruler with golden statues. It's not just a handful of people...

Few people have the balls to do the right thing when the risks pass a certain limit. And yet it's most important to do the right thing at the largest scale.

and when Trump keels over and dies, what happens then? sort of a brittle business plan

The law firms failed to do so.

Some of them have certainly flagged themselves as opportunistic and myopic.

Yeah, specially in STEM.

Despite posturing by some academic administrators, most folks have no social agenda for a country they recently immigrated to.

Imagine actually wanting to work at Amazon

Ahead of the game, soon we'll only be able to spend Bezos Bucks. Company Towns walked so the Company Country could innovate

edit: I've enjoyed two managers since I started working ~2008. The rest were either harmful or generally ineffective, unfortunately. We can get along well... but I know the job's not for me. It's like HR, for the business.

Contrary to GP, career development has yet to meaningfully happen within the same business. I have to move to advance, so I do.

Indie dev turning profit first year is part of the typically success story for indie devs.

It looks very adhoc

So ideally the ml network would solve the problem end to end but these authors seem to be using a network for only one step of their otherwise classic image processing pipeline

What a smear.

A lot of applied work in vision and audio glues together different existing modules, instead of training the whole thing end-to-end.

In an ideal world, things are ideal. But the world isn't a grad student's wet dream.

But there are not as many influential papers in the world as there are people who respond: "Why didn't they train the whole thing end-to-end?"

"Ideally they'd solve the whole problem, but it looks like they just solved one part of the problem"

lol these guys aren’t vcs

Term sheets from vcs increasingly include a “don’t do an ico”

A16z was leading pump and dump schemes in 2020 - 2021. Not sure if they are still doing that in 2025.

No true scotsman fallacy

There are many VCs in the crypto space, and there are many kinds of liquidity arrangements in the crypto space, a subset of them are coin related and snapshots are the preferred way, now. ICO-type ones are pretty rare but the VCs one debut at extremely high valuations for the crypto space

A VC saying “dont do a coin” which is probably what you meant, is also probably not funding a crypto venture. yeah, that makes sense to say you arent investing in a meme coin, since seemingly everyone with a platform sullies their brand this way