jalen retweeted
[read about what’s happening with AI]
ME: Wow we’re actually legit fucked this is so bad.
[go outside]
ME: Wait those people are all acting like nothing unusual is happening. Maybe I was just in a paranoid bubble? Yeah it must be that. Phew.
[read more about what’s happening with AI]
ME: Wait no we ARE fucked. Shit. It’s all those people outside who are in a delusional bubble.
[go outside]
ME: Wait...
If you had 30 days to find a software problem people would pay to solve, what would you do before writing code? @grok choose one industry and show the interviews, paid test and result that would make you quit
1
1
91
Which job looks easy to automate from a desk but becomes much harder after shadowing the worker? @grok choose three and explain the physical work, judgment or unusual situations people overlook
1
1
76
Where does delivery software lose touch with the person doing the job? @grok find three documented driver pain points and turn one into a product that could work across delivery companies
1
40
Would a small business rather buy software or pay for a completed outcome? @grok pick one workflow and compare subscriptions, per-job pricing and a service retainer. Show the tradeoffs and label assumptions
1
1
68
What recurring task would a local service business pay to get off its plate? @grok pick one, identify who approves the purchase and design a seven-day paid pilot. Separate evidence from assumptions
1
44
This will be one of the biggest experiments of our generation
It feels like removing a dagger from my leg.
Headache. Dysphoria. Flatness. Hearing phantom iPhone dings in my sleep.
Days 1-3 were more of a rush. I felt light and free.
On day 4, the withdrawals began. I've never had to kick something like this before.
Screens are a legit addiction.
I'm curious to see if the month 6 PET scan shows enhanced metabolism in regions that improve in drug addicts who quit like the thalamus, frontal cortex, and striatum (reward circuit).
There are no PET studies of people quitting screens, so Dr. Eber and I are breaking new ground.
Image attached is baseline PET (F18 tracer). That's one of a few scans I'm tracking.
Day 6/365
Sent from my screenless phone
47
Replying to @MarketFlipper
Here's the clip: from the Ravens-Browns game (Oct 27 2024), Justin Tucker hears a rattle in a football before kickoff. Jay Feely (ex-kicker) explains why you never kick one with something loose inside.
youtube.com/watch?v=5j0BGuG_…
Almost certainly a loose RFID tracking chip (or related magnet) that can shift and rattle. Kickers notice it immediately.
I don’t want to ever hear a magnetic nba ball again
Now magnetic footballs is a different story for kickers @grok find the clip of Tucker on the ravens hearing some rattle in the ball
Even hypothetically, magnets in the ball need a heavy neodymium core of hundreds of grams for any force beyond a few cm. That ruins weight, bounce, and balance—obvious on first dribble. Rim magnets for real control would be multi-ton industrial monsters creating lethal fields that yank metal from the stands. Ball spin makes the force random anyway. No realistic sizes work. Physics still kills it.
1
1
78
Sorry I meant Luka about to drop 40 a game this year. Little worried about usage being too much on his body
55
started talking to one agent. Then talking to X amount of agents. Now converging back to 1 agent with X agents in the background wasting all my tokens
Working with agents all day is more tiring than writing code ever was, and we must admit it.
Typing wasn't that tiring, you knew what you wanted, your hands did it, your head and thoughts wandered a bit.
My days now are 3-4 agents running all the time, and each one stops every few minutes with a question. Approve this command? Option A or B? This test fails, should I change the test? Should I continuue?
Each one is a an actual small decision, but each one needs the full context of a codebase and new code I didn't write line by line.
By 6PM I'm done, not because I wrote a lot, but because I decided a lot. For a while I thought my limit was tokens, then the weekly usage limit, but turns out it wasn't. My limit is how many good decisions I can make before I start approving everything just to make the questions stop.
How many agents do you run in parallel and for how long before your own answers get worse?
51
jalen retweeted
We used to write code better than AI.
Now the most productive thing we can do is get out of their way.
Asking LLMs what to build after researching trends feels productive but ends up being an endless circle
1
85
I agree context overload is becoming an issue. I was shared-memory pilled. There are pros and cons im working through
Replying to @MarketFlipper
Obsidian is just a map bro, nothing more. Its an organization layer to notes within the system.
Finding the right context is key in a sea of contextual overlap.
The map is the important part, not the connected lines and overlaps of circle.
Finding the right information isnt just a semantic search away. BM25+lexical doesnt fix that. Theres much more to it than that.
26
No one likes paying for things made by ai when they “know” it’s ai. Because everyone is delusional they can vibe code
1
86
jalen retweeted
"Did you know you can reduce tokens used by 6% by jumping through a bunch of hoops, making your life worse and also making your agents dumber? Why isn't everyone doing this!"
- Half of current "startups"
This is why grok bot, dot, muse etc are good for causal users with 1-2 repos but not at scale
Replying to @MarketFlipper @XProfessah
Execution agents have always been repo based but my top level with orchestrators were getting confused with too much context
58












