I don't believe headlines like this. I cannot un-taste remote work. I have come to the pure conclusion that I have been faking, role-playing, acting a whole character at work for years. The morning costume, the morning routine, the persona, the blank stare at the screen ... I'm just tired of pretending.
I'm waking up at 9:30am with bed hair and letting the team know "no updates here" and going back to sleep. You can never ever convince me otherwise that there is a better way to live. Everything will get done, don't call me, I'll call you, see you next week (online). You shall never see me in the real world ever again.
Why would the US security apparatus outsource the model to a private company? DARPA or whatever should be able to finance a frontier model and do whatever they want.
There is a concept of going into the wilderness for some time (as we go through Lent). It's ancient. I wonder if we'll ever find out it's just as useful as intermittent fasting.
Pay attention to the outflow of tech investment in the stock market. That money is going to move into OpenAI and Anthropic IPOs. The valuations will be as big you are thinking because the market believes these companies will represent an entire basket of startups.
I was just looking at the SWE-bench docs and it seems like they use almost an arbitrary form of context engineering (loading in some arbitrary amount of files to saturate context). So in a way, the bench suites test how good a model is with little to no context engineering (I know ... it doesn't need to be said). We may not actually know which models are sensitive to good context-engineering, we're simply assuming all models are. I absolutely agree with you on one thing, there is definitely a ton of low hanging fruit.
Are you certain of this? I know they use a lot of grep to find variables in files (recall reading that on HN), load the lines into into context. There's a lot of common sense context management that's going on.
I'm part of the subset of developers that was not trained in Machine Learning, so I can't actually code up an LLM from scratch (yet). Some of us are already behind with AI. I think not getting involved in the foundational work of building coding agents will only leave more developers left in the dust. We have to know how these things work in and out. I'm only willing to deal with one black box at the moment, and that is the model itself.
Quite frankly, most seasoned developers should be able to write their own Claude Code. You know your own algorithm for how you deal with lines of code, so it's just a matter of converting your own logic. Becoming dependent on Claude Code is a mistake (edit: I might be too heavy handed with this statement). If your coding agent isn't doing what you want, you need to be able to redesign it.
I'm guessing it is for situations like should the Waymo stay in a particular lane or switch lanes, try to overtake another car, etc. That's probably the type of "guidance", which seems a lot like optimization.
I suspect most devices will have cameras and mics on them and will mesh connect as a collective system. OpenAI is most likely working on a suite of devices that would fit this "regalia" of sorts.
I'm waking up at 9:30am with bed hair and letting the team know "no updates here" and going back to sleep. You can never ever convince me otherwise that there is a better way to live. Everything will get done, don't call me, I'll call you, see you next week (online). You shall never see me in the real world ever again.