←
Buck14,215 posts
Buck avatar

Buck

@BucknSF

Slanging bits

◷ Joined March 2021▧ 915 Photos and videos
Buck avatar
Buck@BucknSF·Apr 25

looking forward to the refs hammering the suns today for daring to say they are trash

Chetan Puttagunta avatar
⟳ Buck retweeted
Chetan Puttagunta@chetanp·Apr 25

Given Anthropic's reported limitations on inference, it's pretty straightforward to imagine them competing directly with their API customers for the rest of 2026. OpenAI and Google could gain meaningful ecosystem share.

Buck avatar
Buck@BucknSF·Apr 25

the number of live enterprise agents doing anything complex outside of Claude code/codex is virtually zero. big theme of next year.

The Cognition team released an article describing the challenges of building an in-house AI agent. A big challenge is ensuring your agent can complete real work. Securely and at scale. That requires proper isolation and running agents in specialized machines. The good news: if you run your agents on @e2b, you get over three years of work on this problem out of the box. Hard won lessons and integrated into your agent in an afternoon.

Post attachment
Buck avatar
Buck@BucknSF·Apr 25

NOW exposure to AI growth is just marginal. token spend required to answer a support ticket low and only goes lower.

$NOW CEO on how companies are funding AI: "What I'll tell you is that from a budget perspective, we're seeing it come from a lot of different places, right? It's not one place. A lot of people are funding it because labor budgets are coming down. A lot of people are reallocating technology spend and eliminating more point solutions and really leaning into platform consolidation and platforms like ServiceNow"

Buck avatar
Buck@BucknSF·Apr 25

bull case would be usage patterns evolve dramatically and positively (i.e., code shipped increases 10x). think this is true but bounded by growth potential of corporates. prob could sustain 10k+/user.

Buck avatar
Buck@BucknSF·Apr 25

AI Coding TAM for today's usage patterns. ~ 1k/yr current ASP per user. Power users 10-100x that. But realistically I think 20x ASP potential. 20m devs. So 400b TAM for today's usage pattern. but token costs will depreciate by 90%+ so really 40b

Buck avatar
Buck@BucknSF·Apr 24

the universal narrative: across customers, investors, employees the narrative about a business becomes the same. when you are in a fight for control of your narrative you have to play dirty. be aggressive. guide aggressively into high short interest. over invest. ignore your CFO.

Buck avatar
Buck@BucknSF·Apr 24

I've been FIG hater, but Make-driven acceleration likely in Q1/Q2

Buck@BucknSF7 Aug 2025

you have every lab gunning directly for FIG's market and it trades at 40x sales...sometimes money does fall from the trees.

Buck avatar
Buck@BucknSF·Apr 24

the immediate effect fwiw will be a major step up in revs...just question of curve after that

Buck avatar
Buck@BucknSF·Apr 24

cursor is subsidizing tokens for customers and will stop doing that this summer, because they make no money. will be interesting to see what happens to growth at that point. bull case is it drives demand for their in-house model.

Cursor at -23% gross margin in January $2.7B annualized revenue, up 14x YoY, expects $7B eoy but Claude code caught up fast, hard for cursor to raise xAI only at $3.2B 2025 rev, mostly twitter, really wants Cursors revenue 🍿 to see how it shakes out theinformation.com/articles/…

Post attachment
Buck avatar
Buck@BucknSF·Apr 24

love the Zerg Rush twolves...smoking the Protoss nuggs

Gokul Rajaram avatar
⟳ Buck retweeted
Gokul Rajaram@gokulr·Apr 23

CODING: THE CANARY VERTICAL Coding is the first AI vertical to hit escape velocity, and a preview of every market that follows. (In other words, it's the canary in the coal mine) Coding got there first for a reason: it has the most objective feedback loop in any knowledge work category. Code runs or it doesn't. Tests pass or they fail. That signal richness is why models hit near-superhuman capability on code before law, finance, or medicine. And why developer tools were the first category where AI crossed from "interesting demo" to "I can't work without this." The growth story is well known. The more interesting thing, IMO, is the collision. Anthropic shipped Claude Code. OpenAI has Codex. Google with Gemini Code Assist. Every major lab now competes directly for the developer, with companies like Cursor, @FactoryAI, Cognition and other application layer startups. The infrastructure layer and the application layer are now indistinguishable in software development. Foundation model companies started with model quality but built distribution. Application companies started with distribution and built their own models. Every other vertical will go through the same arc: models get good enough, app companies find the right workflow wrapper, growth takes off, foundation model companies notice and build directly, app companies build their own models for defensibility. Legal is probably 12-18 months behind coding on this curve. Finance is similar. Healthcare is further out, because feedback loops are slower and regulation adds friction. App companies that survive will look like Cursor or Factory: deep workflow integration, proprietary context, switching costs that outlast a model upgrade. Founders building vertical AI companies: the foundation model companies are watching your category's growth curve. They will build what you're building once the market validates it. Become a deeply embedded product that truly runs your customer's business. Build the whole platform so you are irreplaceable before they arrive.

Buck avatar
Buck@BucknSF·Apr 23

the stock down 20% on a meh quarter prob a bit much, but if you believe the AI ACV number you should be asking why they aren't guiding to acceleration.

$NOW Q1'26: Noisy quarter clouds fine results. PT Raised to $226 at Bernstein

Post attachment
NoTechBen avatar
⟳ Buck retweeted
NoTechBen@NoTechBen·Apr 23

Same. It’s a 7 footer grifting with a 6 footer guarding him. It’s just a totally broken culture.

Probably the play that made me the angriest today

0:05