richard kim

cofounder & ceo - jelly ai labs

[richard kim @kimchi_hacks]

Yes and humans are complex and nuanced in their views of what is even considered “good”

So if the quintessential question of what is good re: taste cannot be unanimously agreed upon, why do we think somehow the game is over?

Game getting harder is natural evolution but not the same as game being over.

[richard kim @kimchi_hacks]

@viswa_kotra great my cto will be in contact.

[richard kim @kimchi_hacks]

@imtomcurry @cyntro_py The game is so perverse that you would think we are playing CMBS/MBS trading games circa 2003-2008.

[richard kim @kimchi_hacks]

majority of VCs have a hit rate in finding unicorns at less than 5%.

so by stats avg investor in this space has has heuristics that is no better or worse than luck.

what I hear right now from my vc friends is that deals of $50-500m seed sizes are being done based on where they came from (eg some big research lab) and pedigree.

that’s the anti-heuristic if I ever heard one.

if you think being a top researcher in an ai lab = know how to build a company, recruit, or even build product. Oy vey!

[richard kim @kimchi_hacks]

apples and oranges.

a user who debates with ai who then goes to a site is super high intent by design and did all the comparison (top to bottom funnel).

If an ad shows up at the end of that convo sure & likely true… if not, on some higher level convo prompt -> that’s similar to a google short tail search where conversion rates goes down 70-90% vs a long tail search.

[richard kim @kimchi_hacks]

@MilkRoadAI Relative to what $$$ investment and capital efficiency?

It’s great that investment are coming in and adoption is high but it’s not pari passu.

Apples and oranges.

[richard kim @kimchi_hacks]

@ksws0355460 @JacksAISites At what token economics…
Is the question.

[richard kim @kimchi_hacks]

Every year around this time, I like to review the stuff “X”perts on this platform like to claim and actually see where we are at.

@Nickprince @ChaseMc67 @Lon
So where we at?

[NIK @ns123abc]

🚨Anthropic researcher believes: “maybe as soon as the first half of next year: software engineering is done.”

ITS SO OVER.

[richard kim @kimchi_hacks]

Yassir!

[Yasser @yasser_elsaid_]

If building is easier, build harder things.

[richard kim @kimchi_hacks]

I spent $0.02 using my own skill called common sense.

Stop this madness. Here is my free explanation of Harvey’s growth.

Business Model:
It’s expensive legal tech platform selling to legal firms usually from the top down.

Just think about that for 30 seconds…

How does that selling motion go. You need zero ai skills to deconstruct how anyone would sell an enterprise level platform. Just ask it bing or ask Jeeves or altavista and it’ll still get it right.

You need zero ai or agents to understand the core growth motion of an enterprise legal tech growth motion.

To say funding didn’t help also is highly suspect if you separate sales motion (can’t be captured in marketing traffic) from marketing motion.

[richard kim @kimchi_hacks]

@chandr3w @AllenWangzian Lololol

[richard kim @kimchi_hacks]

@AdvaitRaykar What network effects?
It’s a personal agent.

Is there a multiplayer advantage?

[richard kim @kimchi_hacks]

@haridigresses @alistairmcleay X = amateur pros.

Directionally correct.
Specifically consensus.
Mostly scams.
Rarely gems.

[richard kim @kimchi_hacks]

@market_sleuth Donated. It’s ridiculous how hard most people posting his story make it to even find his go fund me account.

https://t.co/ypIRbrzzOo

[richard kim @kimchi_hacks]

I don’t think that’s the real story under the hood? Macro story is

1) rising rates & Democrats winning and it’s impact to “easy money” / yield requirements

If rates go up, private debt investors need higher yield to get additional funding for growth of data centers. That can only be justified based on payback and Roi. Some data shows Roi is not great which makes people pause on whether or not continued increased demand for data centers will be there

2) LLMs impact near-term to equity vol as they suck in $$$ but no clear path to profitability due to open source question:

markets already priced all of 2028 into perfection so no point in discussing how much is or is not available now. Is pricing and inevitable margins a concern if frontiers ($3 trillion of value) has no real biz model but they promised all this purchase of compute? How will they fund it? What is the real open source challenge here.

3) technicals:

Why can’t Nvidia make new highs if this is so well known. Every donkey knows there’s compute shortage by looking at avail. So obvious that for a $4 trillion dollar biz it’ll get priced in fully by now.

I don’t know the answer tbh just saying this is what I would imagine Wall Street to be digesting.

[richard kim @kimchi_hacks]

@shiri_shh 🔥🔥🔥

[richard kim @kimchi_hacks]

how many of you know what “R” stands for?
this changed the music world forever.

oh Napster days!
@sparker

[richard kim @kimchi_hacks]

if u r reading this comment, pls do urself a favor: stop doomscrolling on X, go outside, and get urself a pain au chocolate.