Appreciation for those of you who follow me on @farcaster_xyz farcaster.xyz/cdt

“I’ll figure it out” has gotten me further than any plan I’ve ever made

Everyone is building AI agents. The difference is what’s behind them. Live from #Links26: @jony_levin introduces the next chapter of Chainalysis — blockchain intelligence agents. Not a new product, not a bolted‑on chatbot, but the evolution of our platform and everything we’ve learned: billions of screened transactions, 10M+ investigations, and more than a decade of blockchain intelligence working alongside your team. Chainalysis Agents are engineered for high‑stakes investigations and compliance around four principles: 🔹 Data quality 🔹 Context & reasoning 🔹 Auditable deterministic workflows 🔹 Humans in control Read the blog to learn more: chainalysis.com/blog/introdu…

been thinking about this. I don't think it's that easy to get balance, because inertia and reputation are just social physics. instead, I've been thinking about "good problems to have". I would rather say yes to most/all things that come my way, and work really hard as being a do-er, than the alternative. but then I work very hard at pattern matching and staying ahead of the curve. common questions/help? that means I need docs/tools/things I can automate. favours? those are good, because you can collect them and use them to help _other_ people when they need them. being "that guy" who people reach out to when they need it? become that guy. I'm constantly thinking of problems that will arise, and doing some work, or even just some thinking, so when the opportunity arises, I can move a little quicker than most. relationships are hard, you got to work on them. yeah I feel swamped and tired a lot, but it's incredibly satisfying. and in a high trust environment, you can then say no and/or ask for some leeway, and people will give it to you because they want to preserve _their_ reputation with you. they'll get it. these are all good problems to have. importantly, this pays off HUGE. you won't see it when you're helping people, or being nice, or whatever. but oh buddy, when the time comes, they'll crawl out of the woodwork to help you, to support you, to defend you, whatever. I need to formalize these thoughts sometime, this was loosely written. but I feel and live it everyday.

My dear front-end developers (and anyone who’s interested in the future of interfaces): I have crawled through depths of hell to bring you, for the foreseeable years, one of the more important foundational pieces of UI engineering (if not in implementation then certainly at least in concept): Fast, accurate and comprehensive userland text measurement algorithm in pure TypeScript, usable for laying out entire web pages without CSS, bypassing DOM measurements and reflow

chat, how do we feel about this for my AI Engineer London keynote? (still can't believe anyone would make me a keynoter).

seems obvious but: things that are changing rapidly: 1. context windows 2. intelligence / ability to reason within context 3. performance on any given benchmark 4. cost per token things that are not changing much: 1. humans 2. human behavior, preferences, affinities 3. tools, integrations, infrastructure 4. single core cpu performance therefore, ngmi: 1. "i found this method to cut 15% context" 2. "our method improves retrieval performance 10% by using hybrid search" 3. "our finetuned model is cheaper than opus at this benchmark" 4. "our harness does this better because we invented this multi agent system" 5. "we're building a memory system" 6. "context graphs" 7. "we trained an in house specialized rl model to improve task performance in X benchmark at Y% cost reduction" wagmi: 1. product/ui 3. customer acquisition 4. integrations 5. fast linting, ci, skills, feedback for agents 6. background agent infra to parallelize more work 7. speed up your agent verification loops 8. training your users, connecting to their systems and working with their data, meeting them where they are

I'm usually not one to write thought pieces without much technical depth. But here we go. Slow the fuck down. mariozechner.at/posts/2026-0…

nah, i think custom ui/generative ui will be the future

code mode: let the code do the talking (aka, after w/i/m/p) wherein I ponder the implications of every user having a little coding buddy, and every "app" being directly programmable on demand. sunilpai.dev/posts/after-wim… lmk what you think.

You suffer through 30 “just paste this into OpenClaw and it fixes everything” posts to find nuggets like this, from one of maybe 10 teams who are actually worth listening to.

networks we live by

/tree is what I think is the coolest feature of pi. I hand over /tree to pi and let it manage the context on its own, which basically replaces compact. github.com/ttttmr/pi-context

Looking for my next thing. Generalist: product, code, ecosystem growth. Co-founded @nounsdao, Prop House & Rounds (grants infra platforms). Started at Coinbase/GDAX in 2017. Interested in stablecoins, DeFi, AI. dms open.

inline table control interaction ⎯⟡°

I've published the first two chapters of a new guide to Agentic Engineering Patterns - coding practices and patterns to help get the best results out of coding agents like Claude Code and OpenAI Codex simonwillison.net/2026/Feb/2…

"now i have the full picture"

here's my file explorer companion for agent uis in the terminal. gives you quick access to copy paths and see changes agents are making across a codebase. it has mouse support, live git status, flat and hierarchical search, theme support, resposive to small widths. works really nicely as a pane next to your claudes or codexes. ❯ bontree . github.com/almonk/bontree