Ofc they can't expose the underlying mythos-class model right. Since they have tightened up the safety classifiers and automated routing guardrails - we are not actually experiencing the actual mythos level intelligence, and our prompts mostly getting routed to opus 4.8. Straight L from anthropic.
US government held back the release of fable and mythos, now them agreeing to the terms and conditions released back again, with restrictions and fallbacks to lower models for some prompts.
GPT-5.6 is released for a small group of 20 trusted organisations, and is waiting for review for a complete rollout.
US government controlling two largest companies at this point.
Would this be seen as starting point of decline of closed source american models and huge boost for chinese models?
We will get to know in near future.
genuine question to all the HRs
how do you do the resume screening of all the candidates, in this era where ones resume is just a variation of the other?
feels filtering by resume is highly ineffective way of picking the right candidates
would be eager to know your thoughts
@samirande_ you may agree. webdev (atleast front end, for now) is heavily automated and AI can do the heavylifting. it will catch with backend probably very soon.
@ramxcodes I personally think it is very much expected whenever a ground breaking technology comes up. The existing systems struggle to get adapt to it instantly. It gets better day by day.
Its good that we don't have to share phone numbers.
But feels like they will go for the blue tick subscription plans in near future. Probably merge it with the current whatsapp plus.
Hope they don't make whatsapp un-usable.
your phone number is personal and sometimes you want to connect without handing it over. that's why we're introducing usernames for WhatsApp.
starting this week, you can reserve a username to use later this year when we launch the feature. It takes just a few seconds, make sure you have the latest version of WhatsApp and then go to Settings > Account > Username.
I have always had a guilt of - this is not where I wanted to be. This is not who I am. Something has gone seriously wrong along the way. Nothing is scarier than avoiding your full potential.
So I have decided to finally wake up.
I am gonna experiment myself for the next 6 months in something I have never tried, always feared to. Starting from absolute ZERO. Forget everything that has happened. All over again.
Goal is simple. Become disgustingly good version of yourself. Be it networking with right passionate people, being up-to-date with this fast paced tech world, maintaining up the fitness, and so well read.
Dec 31, 2026 will be the date.
This is where I document my learnings, efforts, and my thoughts through the whole process.
If you are someone can relate, hopefully you stick around.
We are officially Locking in.
@nia_thinks a aspiring founder of (hopefully) a really good product, building it, by vibecoding, keeping myself at the top of AI news everyday, and loves to explore cool UI designs here. will be happy to connect.
@benln I would not say they are lectures, rather they are more like podcasts, understanding the AI business, promoting their personal company. Personally this can be a good watch. But as "best AI series" where someone can learn to adapt AI, build cool things - I would personally differ.
@_vmlops i would build a solid community first. create one, check the pulse with it, and deploy. its getting way easier to build a MVP than building a community.
@dev_maims most of work i seen on this app are just poc-s. probably mostly due to the fact that they create these for content and engagements, and not for real purpose. they fear if they put it for 1000 of users, the system will break.