every tech executive is talking about making it so anyone on the team can ship code this means engineers focus on guardrails, patterns, etc to allow for this to happen safely but this isn't new! this has always been the job of the senior people on the team, make the less experienced people more productive and you do this by being really good at designing code, and you're gonna have to be really really really good to allow your marketing team to ship changes without things breaking

talking to ceos and eng leaders lately and everyone is running into the exact same problem. as teams grow and devs use ai, they ship so fast that merge conflicts just get out of control. you assign a junior a tiny task like adding a button. ai makes it so easy they do way too much and break things. granular tasks shouldn't exist anymore. in startups devs should just be focused on creativity. put them in their own environment. get them the max cursor and claude plans. let them burn tokens just experimenting with different ui and ux to solve problems. then they just do quick demos for the team. we don't care about the actual code. ONLY the ideas. if an idea is good, an eng leader with deep architecture knowledge just looks at the build. they use ai to write an implementation plan to get that feature built correctly into the main codebase. production code stays PERFECTLY protected by seniors. juniors get to actually invent new features and feel relevant. ceo gets a ton of new product ideas.

it's obvious but it's crazy how much different models can generate differrent output. It's easy when building agents to stick to the same model. But note to self is to always run the flow using many different models in parralel

i use to love subscribing to my fav companies newsletter to see what new stuff they just shipped but now with AI it's so easy to build features or content that I feel the newsletter are just noise. I now mostly just look at top HN posts, YC bookface, or aggregators like TLDR

when building agents, system prompts should always be written by hand. never ask an llm to write or fix it for you. whenever i do that it just ends up polluting the prompt so much. confusing the llm. it might actually be the only thing i still write entirely by hand.

keyboard text replacement is so slept on for AI prompting. i set up shortcuts like "11" for "explain like i'm 5...", "22" for "be concise, ...", etc. those works everywhere, any platform. i just stack them based on what i need.

Making something people love is mostly making something you love and hoping the overlap is real

Another week on the road meeting with a couple dozen IT and AI leaders from large enterprises across banking, media, retail, healthcare, consulting, tech, and sports, to discuss agents in the enterprise. Some quick takeaways: * Clear that we’re moving from chat era of AI to agents that use tools, process data, and start to execute real work in the enterprise. Complementing this, enterprises are often evolving from “let a thousand flowers bloom” approach to adoption to targeted automation efforts applied to specific areas of work and workflow. * Change management still will remain one of the biggest topics for enterprises. Most workflows aren’t setup to just drop agents directly in, and enterprises will need a ton of help to drive these efforts (both internally and from partners). One company has a head of AI in every business unit that roles up to a central team, just to keep all the functions coordinated. * Tokenmaxxing! Most companies operate with very strict OpEx budgets get locked in for the year ahead, so they’re going through very real trade-off discussions right now on how to budget for tokens. One company recently had an idea for a “shark tank” style way of pitching for compute budget. Others are trying to figure out how to ration compute to the best use-cases internally through some hierarchy of needs (my words not theirs). * Fixing fragmented and legacy systems remain a huge priority right now. Most enterprises are dealing with decades of either on-prem systems or systems they moved to the cloud but that still haven’t been modernized in any meaningful way. This means agents can’t easily tap into these data sources in a unified way yet, so companies are focused on how they modernize these. * Most companies are *not* talking about replacing jobs due to agents. The major use-cases for agents are things that the company wasn’t able to do before or couldn’t prioritize. Software upgrades, automating back office processes that were constraining other workflows, processing large amounts of documents to get new business or client insights, and so on. More emphasis on ways to make money vs. cut costs. * Headless software dominated my conversations. Enterprises need to be able to ensure all of their software works across any set of agents they choose. They will kick out vendors that don’t make this technically or economically easy. * Clear sense that it can be hard to standardize on anything right now given how fast things are moving. Blessing and a curse of the innovation curve right now - no one wants to get stuck in a paradigm that locks them into the wrong architecture. One other result of this is that companies realize they’re in a multi-agent world, which means that interoperability becomes paramount across systems. * Unanimous sense that everyone is working more than ever before. AI is not causing anyone to do less work right now, and similar to Silicon Valley people feel their teams are the busiest they’ve ever been. One final meta observation not called out explicitly. It seems that despite Silicon Valley’s sense that AI has made hard things easy, the most powerful ways to use agents is more “technical” than prior eras of software. Skills, MCP, CLIs, etc. may be simple concepts for tech, but in the real world these are all esoteric concepts that will require technical people to help bring to life in the enterprise. This both means diffusion will take real work and time, but also everyone’s estimation of engineering jobs is totally off. Engineers may not be “writing” software, but they will certainly be the ones to setup and operate the systems that actually automate most work in the enterprise.

it’s crazy to me that people still spend so much time on slide presentations. Just give me the information in a big text not formatted i don’t mind. i’ll always put it in an LLM anyway so i can consume it how I WANT

many people have been talking about it but it’s crazy how now with AI I end up coding much more than before . i’m like working on 5 different features and 3 different projects at the same time and i feel anxious if i stop having this velocity. like if i take a pause and my agents aren’t working for example. forcing myself to go for a walk is now SO beneficial. clear my brain. take a step back to see direction i’m going with everything etc. it was always good to take a break but i feel now even more.

wild how people confidently state "facts" now and you can just tell an LLM hallucinated it lol

one thing i always asked myself when travelling was always why some places would try to make it a good environment for nomads. Like we’re super cheap, it’s not us that will help their tourist economy. We stay in one cafe all day buying almost nothing. Try to do all the hacks so spend like a local etc

To quote from my keynote at Vercel's internal offsite: Software is free as in puppies. It will pee in your bedroom and eat your furniture. The weight of every line of code is real. We will need to maintain it. We will need to port it. It goes into the context window. And somebody in this room will get paged at 2am because it did something unexpected

thinking opus 4.6 pricing is a total marketing play. jack the price to squeeze cursor, then drop $50 credit to bait people into claude code. wild that $50 only buys a few messages. smart move to make us think we're getting free cash while cursor panics

building and trying new things have become so fast and easy that i’m sure i’m repeating the same experiment without noticing

everyone reads a title and thinks they know it all. it’s crazy. we only know a tiny bit but talk like we’re experts. we’re all just repeating 1% of the truth and it’s wild (i do this all the time too)

with code getting so cheap, everyone's launching way too fast. impossible to keep up. companies always built too much bloat but now you gotta build like 10x less than before just to stay sane

i feel travel is overrated if you have a base. cause then in bad moments you’ll always think of how comfortable you were back home

Making devs finally write docs by calling them “skills” is a great arb

just use some “older” models. create as many sub agents as you want. even run the same one 3 times in parallel and have another one merge the results if you need to. this will still be so much cheaper then newer ones and you’ll have better and more deterministic results