I too was puzzled by the response from Claude.
I am using the Anthropic workbench with claude-3-5-sonnet-20241022 (latest)
But it think it has to do more with the freshness of training data.
AWS IPV6 Egress is a new technology from AWS which was introduced only recently. Previously, we had to deploy NAT gateway which supported IPV4. I am assuming claude-3-5-sonnet-20241022 (latest) was not trained on this data.
Yes. I find it a bit funny how much people care about leaderboards. I see models going up and down, winning this or that benchmark and yet, for me, Sonnet 3.5 still beats the crap out of all of them.
Not bioavailability, but rather which amino acids and the ratio of amino acids. If they're not balanced out then one of them becomes the limiting factor. Most modern plant based protein supplements are a blend of proteins that are in balance to avoid that problem, for instance rice and pea blends.
Soy is actually pretty good, very comparable to meat.
Say I had rice for breakfast and peas for lunch - will this be balanced? What if I had them together? Do I need to literally blend them together? Or is it only when they are grown as a same cell it is actually complete protein?
Very probably. The duration scale your body assimilate amino acids together is about 12h-36h IIRC. Most plant eaters I know don’t bother counting anything at all, and the natural craving helps: if I eat a full meal of rice, the peas will be more appealing for the next meal than another bowl of rice. But for taste and texture it’s often better to mix many things in the same meal.
the 'data' used is pretty suspicious in this article
obviously, international founders moving to america will have a better time raising money
and if you're selling B2B SaaS and 90% of your customers are physically headquartered in SF, you're handicapping yourself living elsewhere.. at some level people will just ignore your sales motion if they don't see a san francisco headquarters on the website
i've lived in both SF and NYC and worked on my own startup in both
NYC has a pretty vibrant startup ecosystem but doesn't compare with SF, except for some reason there's way more healthcare founders in NYC (prob bc bay area healthcare scene is dominated by large hospital systems that are impossible to sell to)
but SF is such a disgusting city overrun with homeless people and crime
NYC isn't exactly the prettiest either but pick your poison i guess
i personally prefer south bay over SF
hope SF gets its act together since I don't feel safe living there
Seems like the author is a slow metabolizer like me. Interestingly enough, it seems that the caffeine half life can be shortened by eating broccoli (https://pubmed.ncbi.nlm.nih.gov/17266520/) - by having coffee in the mornings and broccoli with dinner I've been able to enjoy the mood boosting effects of caffeine during the day and still be able to sleep fine.
Sleep quality aside though, I thought my anxiety was due to stressful job, but maybe it's due to all the caffeine I have in an attempt to be productive.. perhaps something to experiment with myself.
A lot of the back and forth in this thread seems to be overlooking the glaring fact that caffeine's half-life varies from 2 to 12 hours depending on the person. Getting to 90% elimination could take one person ~6 hours and another person ~36 hours.
This, on top of the fact that caffeine has been shown to increase cortisol secretion and adrenaline, as well as boosting the effects of dopamine and norepinephrine in the brain, should give a very clear indication that caffeine is going to have drastically different effects from person to person.
Anyone who is struggling with sleep, anxiety, and/or mood disturbances should consider reducing or eliminating caffeine for several weeks to see if it helps. Don't expect it to be a magic bullet to solve all problems, just like anything else. Bodies are complex, and when we are chronically suffering from sub-optimal functioning, it's a good bet that their is some complicated and highly variable biochemistry going on that's keeping the body from homeostasis / optimum function.
