Su Park

terminal bench · post-training & evals
against global banality
prev @ltiatcmu @columbia

[Su Park @sunotsue_]

meanwhile at the warehouse / sf finally stopped being lame as of 2026

[Deedy @deedydas]

The vibes in SF feel pretty frenetic right now. The divide in outcomes is the worst I've ever seen.

Over the last 5yrs, a group of ~10k people - employees at Anthropic, OpenAI, xAI, Nvidia, Meta TBD, founders - have hit retirement wealth of well above $20M (back of the envelope AI estimation).

Everyone outside that group feels like they can work their well-paying (but <$500k) job for their whole life and never get there.

Worse yet, layoffs are in full swing. Many software engineers feel like their life's skill is no longer useful. The day to day role of most jobs has changed overnight with AI.

As a result,
1. The corporate ladder looks like the wrong building to climb.
Everyone's trying to align with a new set of career "paths": should I be a founder? Is it too late to join Anthropic / OpenAI? should I get into AI? what company stock will 10x next? People are demanding higher salaries and switching jobs more and more.

2. There’s a deep malaise about work (and its future).
Why even work at all for “peanuts”? Will my job even exist in a few years? Many feel helpless. You hear the “permanent underclass” conversation a lot, esp from young people. It's hard to focus on doing good work when you think "man, if I joined Anthropic 2yrs ago, I could retire"

3. The mid to late middle managers feel paralyzed.
Many have families and don't feel like they have the energy or network to just "start a company". They don't particularly have any AI skills. They see the writing on the wall: middle management is being hollowed out in many companies.

4. The rich aren’t particularly happy either.
No one is shedding tears for them (and rightfully so). But those who have "made it" experience a profound lack of purpose too. Some have gone from <$150k to >$50M in a few years with no ramp. It flips your life plans upside down. For some, comparison is the thief of joy. For some, they escape to NYC to "live life". For others still, they start companies "just cuz", often to win status points. They never imagined that by age 30, they'd be set. I once asked a post-economic founder friend why they didn't just sell the co and they said "and do what? right now, everyone wants to talk to me. if i sell, I will only have money."

I understand that many reading this scoff at the champagne problems of the valley. Society is warped in this tech bubble. What is often well-off anywhere else in the world is bang average here.

Unlike many other places, tenure, intelligence and hard work can be loosely correlated with outcomes in the Bay. Living through a societally transformative gold rush in that environment can be paralyzing. "Am I in the right place? Should I move? Is there time still left? Am I gonna make it?" It psychologically torments many who have moved here in search of "success".

Ironically, a frequent side effect of this torment is to spin up the very products making everyone rich in hopes that you too can vibecode your path to economic enlightenment.

[Su Park @sunotsue_]

@andyfang v tempted to build a Terminal Bench task around this — multi-objective optimization over daily meal credits, delivery windows, dietary constraints, coupons, dynamic inventory, etc

[Su Park @sunotsue_]

@hyerijo_ how respectful

[Su Park @sunotsue_]

@lily_gpupoor telco is oversaturated at this point

[Su Park @sunotsue_]

@thkostolansky phil night this wk!

[Su Park @sunotsue_]

ily seoul 🩷 my cutie home city that never stops reinventing itself. i grew up spending weekends at coex, and coming back for #icml last week felt both strange and familiar. now back to soulless sf 🥀

늘 조금 낯설고 제법 따뜻하고 꽤 정갈한 내 도시!

[Su Park @sunotsue_]

@m_wulfmeier real

[Su Park @sunotsue_]

data’s always been the only constant

[Logan Kilpatrick @OfficialLoganK]

it’s surprising to me how many people seem to not understand that great models are built with super high quality curated data

finding novel ways to create / get this data is a huge edge

[Su Park @sunotsue_]

@spencermateega @AfterQuery 🚀📈

[Su Park @sunotsue_]

@CosmicDude3000 love the hanoks surrounding it too

[Su Park @sunotsue_]

before flying back to sf, do find time to check out korean modern art! art seems to be the one thing sf struggles to cultivate despite all its pursuit of science and progress.

my personal favorite seoul art museums, in order:

1. amorepacific museum of art (the current lee bul & nam june paik exhibition is worth it)
2. mmca (right by the palace, in the heart of samcheong-dong)
3. leeum (run by samsung in hannam)

[Su Park @sunotsue_]

back home in seoul city for #ICML 🇰🇷

I’m Su, based in SF, making TerminalBench tasks by day.

let’s grab coffee if you’ve been thinking deeply about data efficiency, data attribution, or simulation lately - I’ll take you to the cutest ones (I grew up here!)

here till monday!

[Su Park @sunotsue_]

I think this conflates data efficiency with information efficiency? better attribution, curricula, synthetic data, and feedback signals can dramatically change what the model learns, even in low-resource settings.

the objective isn’t just fewer tokens, it’s higher information density

[Su Park @sunotsue_]

@oneill_c would love to beta test!

[Su Park @sunotsue_]

iteratively hillclimbing an artificial wall, v5 over v2 unequivocally suggests gains in long-horizon endurance

[Stella Biderman @ ICML @BlancheMinerva]

I’m walking back from an AI event to my hotel and I look up and see a rock climbing gym called “Benchmark.”

Is this what living in SF is like?

[Su Park @sunotsue_]

@m1nj12 someone needs to make a nangmyeon map in english

[Su Park @sunotsue_]

@himanshustwts will bci break out of its hype cycle this time

[Su Park @sunotsue_]

@imempowa @tiatplace hiii i loved the cybernetics talk &lt;3

[Hamish Ivison @hamishivi]

Trained some terminal agents with friends!

Introducing Tmax, open RL terminal agent models. Under default settings and shorter length (65k) token budgets, tmax outperforms prior open work on terminal use. We are releasing all data+weights+rollouts publically!

[Su Park @sunotsue_]

@ShaiviRau @UseLitmus @ycombinator riverside &lt;3

[Giang Nguyen @giangnguyen2412]

https://t.co/Y3Wp4WnyYp