


Working on @tempo. Previously at Farcaster and Coinbase.


two use cases for agentic payments are emerging - commerce and services. commerce is the one most people think of when they hear agentic payments. it's the buy me a movie ticket use case, which agents can do with cards. but we can improve on two things. the first is a deterministic handshake for the payment. agents computer use their way through browser forms and card auth which is slow and failure prone. the second is flexible auth. cards often require real-time approval meaning you can't tell your agent to pay in the future on your behalf. it breaks the "buy this as soon as its available" use case. service use is less obvious but possibly more important. it's enabling your agent to purchase services quickly to achieve high level outcomes. it's when you say "make me a movie" or "find me the head of sales at x" and the agent needs to purchase some storage, compute or data to achieve a better outcome. today humans create accounts and feed them to agents which is cumbersome, especially when the best api for something might change every few months. a service that made it easy for agents to sign up on behalf of a human and pay-per-use will become instantly more legible to agents.

agents are sloppy writers but good critics i've found they can give useful feedback if you give them a sample of posts you like and ask them to distill a style guide to critique future writing.

mpp is bringing identity to agentic payments. services can check that an agent is operated by a human who owns an email address, api key or has a proof of human verification. the protocol supports flexible, in-line authentication so you can even define custom auth types.
I | Introducing expanded identity primitives for MPP - supporting any identity protocol for machine payments

some thoughts on writing software in nov 2026: - it feels like magic - the first 80% is 100x faster - rewriting systems is also 100x faster - opus 5.5 is very good at a lot of things - "deep flow state" is much easier to reach - almost anything can be built in typescript and rust - building secure things is very, very hard the optimization function is how to maximize agents working on goal with the fewest interruptions needed. some things that help: - know where to trust vs guide (and keep retesting) - give them agency like "i'm going to bed, get this done" - if there is a reward function, ask them to hillclimb - the socratic method works wonders - skills and memories can often hinder - asking agents to review themselves still helps - cowriting a plan/spec/invariants helps sometimes managing your own psychology is equally important: - be a lot more ambitious about what you can achieve - avoid dopamine traps / false signals of progress - get good at parallelizing and context switching there are also some things that don't quite work fluidly yet, but will get unlocked in the very near future: - software factory or "find and fix the bugs" - multiplayer agentic workflows - agents managing other agents for high level outcomes - organizational memory

increasingly more pilled on swarm riding after watching @gakonst work there is a massive unlock in velocity if you can transmute your requests from 'build me x' into 'create the system that solves y'. the part that requires skill is 1) defining y correctly and 2) understanding the limits of the frontier and not getting too derailed by false progress
Swarm riders and Swarm-native companies AI-use even among highly AI-native people is power-law distributed. There seems to be a new class of people — call them “swarm riders” — who are capable of consistently commanding large swarms of agents to productive ends. This is a new kind of managerial skill, analogous to the skill it takes to manage large organizations of people, but wholly different in nature and of course more technical. Based on the swarm riders we know (including many at @Tempo and @Paradigm), these people are capable of productive token spend that is 1000x-10000x higher than median software engineers, and 100x higher than top 5% Claude Code users. The art of swarm riding is different from software engineering and more akin to machine learning, where practitioners grapple with black box systems: defining reward functions and evals, transmuting non-verifiable problems into verifiable problems, running experiments, etc. Swarm-native companies are a new class of company founded by and populated by swarm riders. These companies are organized around swarms in a way that conventional companies will struggle to replicate, as the cost of internal human coordination is sand in the gears of an effective swarm. Effective swarm-native companies will define the right reward functions and evals, remove functional and bureaucratic barriers, and then let the swarms cook. We’ve gotten to see a few swarm-native companies up close — reshaping financial rails @Tempo, building a new gaming platform @Spawn, cyber swarms at Cathedral Systems — and it’s clear these teams wield a significant advantage over pre-AI teams. Some questions I’ve been wondering about: 1) How fast will swarm riding diffuse through the economy? The implications for token spend could be significant. 2) Can swarms be effectively adopted by existing companies or will it generally track the formation of new swarm-native companies? 3) What becomes scarce as swarm capabilities themselves commoditize (through frontier models becoming better at swarming, swarm capabilities built into harnesses, etc)? Cultures and teams that can define reward functions well and otherwise get out of the way?

A very generous gift from Brian! California needs more of the energy that Daniel beings —embracing tech, encouraging and celebrating public giving. It’s sad that so many politicians focus on vilifying success instead
Today, I stood with Airbnb co-founder and CEO Brian Chesky to announce his $100 million gift to San Francisco’s parks. This is the largest single-donor gift to the city in San Francisco’s history, and will support the development of new parks, upgrades to existing spaces, expanded hours at recreation facilities, and much more. An initial $39 million will help unlock five projects in neighborhoods across our city. SoMa, the Mission, Bayview-Hunters Point, Potrero Hill, and downtown will all see transformational projects finish sooner, bringing more welcoming public spaces to residents across San Francisco. As Mayor, one of my top priorities is making San Francisco a city where families can put down roots and build their futures. Earlier this year, we launched our Family Opportunity Agenda—a citywide effort to bring down the cost of living and expand economic mobility for San Francisco families. We're building more housing, protecting tenants, expanding access to affordable childcare, and keeping public transportation safe, reliable, and affordable. We're also investing in the public spaces families rely on every day. Our parks are an essential part of what makes San Francisco such a special place to raise a family. Our recreation programs provide exceptional and affordable opportunities for San Francisco families. With this investment, we're bringing those opportunities to more neighborhoods across our city, expanding access to recreation, and making our parks even better for the families who use them. San Francisco has created enormous opportunity for the people who have built their businesses here. And with that success comes a responsibility to give back to the city that helped make it possible. So, my ask to every leader who sees what Brian has done today is simple: help us create more opportunities for the people who call San Francisco home. Now is the time. Our city is coming back. Don’t just watch it from the sidelines. Be a part of it.

All kinds of businesses borrow against collateral to stay operationally efficient. These credit markets move trillions of dollars, but are illiquid, opaque and expensive. @tempo is bringing credit onchain so borrowers and lenders can connect directly and efficiently.
1/ We’re launching credit on @tempo. Businesses can borrow against their payments and receivables from a broad market of lenders, with pricing and terms that fit how they operate.

managers looking at everyone whose token spend drops after 4pm PT today

i've been using opus 5.5 heavily this month, and it feels like the first step function since my "claude code" moment last year. it is now possible to design and create delightful software and games even if you have absolutely no technical skills whatsoever.

Building the first 80% has gotten 100x faster. Building the last 20% has gotten maybe 5x faster. You can guess whats unfolding in every company that doesn’t have a product dictator.

smart devices are going to be a nightmare to manage. millions of tvs and fridges that can listen in to what you are doing in your home, with backdoors waiting to be blown wide open. disconnect them from the internet while you still control them.
BREAKING: This morning’s cyberattack against supermarket fridges and freezers has now spread to Belgium, Sweden, Luxembourg and Estonia. The hackers increased the temperature up to 50C. Food worth millions of euros has been spoiled. Experts warn that such cyberattacks on supply chains and logistics can constitute a threat to national food security in many Western states.

the disruption of software is probably the gentlest way that ai could have entered the world. if any other industry was changed as much within a year, there'd be striking unions and turmoil everywhere.

they let this madman build frames again
this is my job now

👀👀
@nanocodexyz make a shared tap counter

simulatability appears to be the best predictor of how quickly ai will replace humans in a given domain. math can be simulated perfectly and agents are producing remarkable results on auto-pilot. software still needs business context, so agents are mostly on auto-pilot but still need human steering. hardware is a click above this, given slow feedback loops with manufacturing today, but it's likely a matter of time before we see the first agentic factories that are hyper tuned to manufacturing end to end. biology, on the other hand, seems infinitely harder to simulate given the quantum complexity of modeling cellular behaviors. is agentic acceleration on the field going to be measured in percentages and not orders of magnitude given this complexity?