fooster

Sure, but there are less assertions to check. The surface area and boundaries are smaller and more limited.

So less meticulous care then?

I don’t find that to be true?

I often have many active. Bug investigations, writing code, reviewing, checking logs after deploys and so on.

Cannot with our current level of technology. You are not going to provide the required level of power in Canada during the winter with wind or solar with todays battery technology.

I asked Claude:

"If combined wind+solar output drops to ~10% of nameplate during one of these (a standard threshold), a ~77 GW fleet sized to meet average winter demand produces ~7.7 GW against a ~22 GW cold-snap peak — a 14 GW shortfall that storage alone has to cover. That works out to roughly 340 GWh for a 1-day lull, ~1 TWh for 3 days, ~1.7 TWh for 5 days, ~2.4 TWh for a week, and ~3.4 TWh for 10 days. Ontario's entire current and under-construction battery fleet sits in the single-digit GWh range, so even a mild 3-day lull needs ~100-200x what's actually being built, and a serious week-plus event needs 400-600x — which is why lithium-ion batteries work fine for hourly duration but make no economic sense at the multi-day scale these lulls demand."

Another other things nuclear power plants don't take 15-20 to build in sensible economies. You also cannot use wind & solar + batteries to replace nuclear power.

buy glm 5.2 subscription and enjoy? and for the same money you get way more usage with glm?

ollama cloud, neuralwatt.

I don't think it is fair to say that opus or gpt 5.5 are subsidized? inference for both anthropic and openai are very profitable.

There are lots of subscription plans with acccess to GLM 5.2.

It is neither cheaper or easier to get this build?

"amateur crank claiming"

I don't know why you want to stoop to name calling which violates the guidelines and the spirit of this site.

"without any critical review" is also seemingly untrue: the post says Rutgers and Cambridge are reviewing it

Alot of the comments in this thread are disappointing. Rather that celebrating an achievement (whether or it is validated yet), many of you seem to want to put him down, or make it seem like claude did all the work.

Claiming that claude did all the work is patently ridiculous. Claude is a tool, like any other. The corpus of linear A is ~7500 characters across ~1500 inscriptions and claude, no matter how smart, doesn't just solve that on its own.

What a shame.

The number of doomers and luddites with extremely hyperbolic views on a site for tech entrepreneurs is wild.

Yeah, that totally didn't happen the majority of the time.

Build a more specific skill the for the exact workflow you want?

The mcp support gets more support because there is no cli.

You are trying to make friends not interview people. The easiest way to make new friends is to engage in a common activity. Sports, hobbies, music, clubs whatever. Join one and see what happens.

He’s not telling the truth.

https://economics.td.com/ca-canada-us-trade-balance

“for you”. Not because you were asked mind you.

Appeasement does not work. It emboldens fascist dictators.

You must be new. github was never that stable.

The unfortunate reality is that unless you use these tools it is impossible to keep up. People using these tools well are substantially more productive, and often they were already the most productive.

by and large yes.

All of that aside it is not American territory and it won’t be unless it is taken by force against the will of the people that live there (and Denmark). People don’t like be threatened in particular by countries which they’re allied with.

This might have something to do with the US president threatening to attack Greenland? Decoupling from the US is the only sane thing to do.

codex. codex is also pretty garbage compared with claude code. The permissioning system in claude code with auto mode is now pretty fantastic. With codex the only vaugely usable mode is yolo mode which is bad for obvious reasons.

The problem I ran into was, using the workflow I use with claude, the code that being written wasn't good, missing edged cases, incomplete.

After reviewing the code, I also found it was annoying to get GPT 5.4 to actually fix the code based on my prompts compared with opus. I had to be far more specific and direct (which is related then to missing edge cases, complete, etc).

I found GPT 5.4 terrible. I just tested 5.5 and compared with opus its still not great.

Sorry you are missing out. I use claude all day every day with max and what people are reporting here has not been my experience. My current usage is 16% and it resets Thursday.