TrackAndTravel

Adtech // Racing // Web3

[TrackAndTravel @TrackAndTravel]

@apralky You know better than to make a trait assumption from a distribution's single longest tail outlier. I imagine wealth within reasonable human function has some selection here.

[Will Manidis @WillManidis]

https://t.co/aHOHtb1rcW

[TrackAndTravel @TrackAndTravel]

@sheikhswampert I understand the take for the original ending, there was reasons for that (anno depression, budget issue).

But EoE was brilliant and if it was incoherent to you that's a skill issue.

[TrackAndTravel @TrackAndTravel]

@apralky But this very thing has always puzzled me as well. I've had relatively intelligent 1SD+ folks I went to school with drink and play LoL with friends and push to Gold without taking it too seriously. Hard stuck bronze 800 games is indicative of a trait far below average.

[TrackAndTravel @TrackAndTravel]

@apralky Former diamond player: this thought experiment would be easier to explore with game footage. There are a number of variables in the game that drive success and taking note which ones he's worst at narrows this down. Micro, Macro, tilt, reaction time etc.

[TrackAndTravel @TrackAndTravel]

@punished_daniel @apralky Yes, but I think his point is in the event of how statistically unlikely this is, there might be an additional variable that accounts for that statistical improbability.

The odds favor an element in play that may not be immediately apparent or easily measurable.

[TrackAndTravel @TrackAndTravel]

@forloopcodes This.

People have been overcomplicating this stuff. Automate your tasks with .skills using an agentic AI one at a time. Upmodel the config process as you figure out the integrations, scripts and calls necessary. Then run a lighter model for the actual execution. Example below

[TrackAndTravel @TrackAndTravel]

My suggestion to anyone in a white collar career:

Replicate yourself and your tasks one .skill at a time. Start with the easy stuff. Do the hard stuff later.

[forloop @forloopcodes]

2026 ai bubble is peaking itself right now: someone just dropped a new benchmark on claude skills and the result of their paper is actually insane

the paper explicitly states smaller models with skills beat larger models without them

a smaller model like claude 4.5 haiku equipped with high quality skills smokes a raw state of the art opus 4.5 model by about 6 percent (27.7 vs 22.0)

imagine getting sota level performance from a free model, its basically cheating, you just have to manually spoonfeed it a basic markdown file explaining how to do its job

all of you opus guys are dumb, you can literally spam haiku with skills and get things shipped in 5x lesser time and 0 cost

even wilder thing is that codex gpt 5.2 fails on the pareto frontier. codex burned massive compute and costs, just to get completely mogged by gemini 3 flash hitting maximum performance at a fraction of cash

i can believe skill engineering is now a valid, mathematically proven substitute for compute

over that, it says self generated skills provide zero benefit on average and show negative deltas on 16/84 tasks. if you give an agent more than 3 skills at once, it bloats its context and completely fails

[TrackAndTravel @TrackAndTravel]

@hasantoxr Paper is from May 9th - low quality bait.

[Boring_Business @BoringBiz_]

How life feels when you never learned to code and it is the first thing being automated by AI

[TrackAndTravel @TrackAndTravel]

@PalmerLuckey If you ever make a Game Boy Advance Mod Retro, I'll pre-order first day. Lots of Advance titles I'd love to revisit with this type of experience.

[TrackAndTravel @TrackAndTravel]

@bumbadum14 Add club race tracks to this list. An extreme example of this phenomenon.

[TrackAndTravel @TrackAndTravel]

@PDmarks @shefska SHEEEEEESH

[TrackAndTravel @TrackAndTravel]

@chrissyxchi Me and @palmtoptigah w/ her coloring books

[Tyler Winklevoss @tyler]

Let's all be clear on one thing. @GaryGensler is evil. He should never again have a position of influence, power, or consequence. Any company, university, or organization that hires or works with him post-SEC is betraying the crypto industry and should be boycotted aggressively. This the only way to prevent this kind of misuse of government power from happening again in the future.

Gensler’s behavior can't be explained away as a good faith mistakes. It was entirely thought out, intentional, and purposeful to fulfill his personal, political agenda at any cost. Even if this meant nuking an industry, tens of thousands of jobs, people’s livelihoods, billions of invested capital, and more. Ironically, his sociopathic ambition ended up torching his own political party.

No amount of apology can undo the damage he has done to our industry and our country. This type of person has no place at any institution, big or small. Americans have had enough of their tax dollars going towards a government that is supposed to protect them, but instead is wielded against them by politicians looking to advance their careers. It's time for this pathology to be stopped once and for all. Onward!

[Ross Ulbricht @RealRossU]

Immense gratitude to everyone who voted for President Trump on my behalf. I trust him to honor his pledge and give me a second chance.

After 11+ years in darkness, I can finally see the light of freedom at the end of the tunnel. Thank you so much, @realDonaldTrump 🙏

[Carnivore Aurelius ©🥩 ☀️🦙 @AlpacaAurelius]

It's time everyone admits the bitcoiners were right

[litquidity @litcapital]

When your buddy is saying they won’t buy more bitcoin at $89,000 because it feels too risky

[Ash Crypto @AshCrypto]

If you have survived

- BTC crash from $69k to $15.4k
- LUNA to $0 by Do kwon
- 3AC going from $18bln to $0
- Celsius and voyager bankrupt
- FTX / FTT Meltdown
- USDC de-peg
- Binance bank run
- CZ stepping down
- SEC suing everyone
- 2 year bear market

You deserve this pump.

[TrackAndTravel @TrackAndTravel]

@dgtl_ronin W