top skills in the AI world
1 - mental model / flow: ability to hyper focus on one big task and let agents windows do small pebbles along the way that you easily review quickly
2 - AI harness: very important, heavily customize your local agent system beyond skills and prompts
I spent $8,700 on first dates after moving to NY a year and a half ago. Here is what I learned.
50 first dates, 10 second dates, 1 girlfriend (we've been together for 4 months now). All from Hinge. An average of $174 per date.
Looking back I could have done a better job at filtering. Goal was always a long term relationship. Best filter was a call before the date (highly recommend that actually). Can save both of you a lot of time. And sets the expectations right.
All women were between 25-32. I'm 30 years old. You can definitely find the love of your life on the apps. I spent most of my time in the office so its very hard for me to meet women irl. I'm also not really a bars/club guy.
I was looking for a very specific person. Similar cultural/religious background. Just get Hinge pro or whatever its called. Makes it easier to filter and gives you a boost.
Kinda hard to say but true. If you are a man below 50% in looks/status or whatever you want to call it you are going to have a hard time. there are just more men than women on the apps (see the report). Set your expectations accordingly.
The height thing is true. I'm 6'1 so it was not a problem for me. But heard this complaint from every single guy under 6 feet.
Also dating is expensive. Especially in NY. Maybe just do drinks or something. I always did dinners and insisted on paying for them (maybe an egyptian thing).
Out of the 50 first dates, I probably liked 20. 10 liked me back. One girl I dated for 3 months but it turned out it was not a long term match. Great girl though!
I was very lucky I guess but I never had an awkward experience (definitely more risky for a girl). Actually had some really interesting conversations even if we both knew this would not be a romantic match.
kinda weird to say but online dating is a funnel. you have to meet people to figure out what you like and what you dont like.
took me around 50 tries to find my current girlfriend. you can see the full report in the tweet below.
Longevity of 100 years with 0 offspring = only 100y of actual longevity.
60y longevity w 10 sons that live 60y each = 60y + (10 x 60y) = 660y, excl. grandsons.
This is the math of the offspringmaxxer.
Only a man who loves his genetic legacy more than himself thinks like this.
this is very interesting - i remember automating banking processes of filling forms in outdated windows XP systems in Brazil 10 years ago using a Java library that matched pixels for clicking around in things, way before any AI
this is like the supercharged version of that
Today we're bringing Cua Driver to Windows: background computer-use for any agent. Claude Code, Codex, or your own loop can drive real Windows apps through CLI or MCP while your desktop stays usable, with true multi synthetic pointer support.
1/6
really exciting demo and i believe thats the goal of it eventually -- its just hard problem to crack
immense potential tho IMO just need a couple of passionate people working on it and taking it thru the finish line
this is really exciting but this should be the mentorship UX rather than the productivity UX
@heyclicky needs to deeply embed into OS and browsers so it can execute things faster (and maybe add some small LLM fine tuned for every environment: one for OS, one for browser, etc) by taking shortcuts like doing direct requests bypassing UI and then just taking user to the end screen directly as if saying "see i did it all"
e.g. needs to double triple bet on AI harness on background jobs for specific environments so it can abstract all the noise and do things faster than an human can do and then clicky can live to be that always on assistant that while you are writing your one pager for the next product at the company it is buying you groceries on amazon at the speed of light