Every claim below was made in this piece, quoted word for word and numbered in the order the piece makes them, so you can read it there rather than take our word for it. The sentence above each quote is our reading of the claim, not their wording. Each quote was checked against a stored copy of the page at build time; where the two differ, the quote is the fact.
Their wordsUse the latest and greatest models. Everything else is potentially a waste of time. If a model is wrong 10% of the time, you don't know which 10%, so you end up making multiple attempts.
Their wordsI'm not convinced it adds slop. And even if it does, I'm not convinced the result is unmaintainable.
Our reading
Tell a coding agent what you want rather than how to do it; you will often get better results.
Their wordsDescribe What you want, now How you'd do it. You'll often get better results.
Their wordsWhat stopped hallucinations? Tool calls. They let agents touch our reality. If the agent can't see something, it makes it up.
Their wordsSo let it run what it builds end to end, let it read production logs, deploy, watch the rollout, cut the release, flip the flag, triage the bug report and email the customer. Anything you can see, the agent should see too.
Their wordsKnowledge piles up in the codebase: AGENTS.md files, skills, scripts, docs. But do agents actually use it? Spawn five blind agents, control vs. treatment, and find out. Agents are really, really good at running experiments.
Their wordsEach agent should be able to install and run whatever it needs to get job done. Easy in remote machines, yucky in local dev.
Their wordsDon't iterate with the agent. Let it build tools to iterate. Not "let's try this. ... ah no, let's try that." Instead: "Generate 20 variations and a little page that lets me combine and test them."
Their wordsAsk for irrefutable proof. Biggest game changer. Agents are really good at producing proof: end-to-end walkthroughs, demo videos, before/after numbers.
Our reading
Don't wait for your agents; let them wait on you, with bigger, batched, asynchronous tasks.
Their wordsDon't wait for your agents. Let them wait on you. If you worry about TTFT, you're probably doing something wrong. Lean into async. Give them bigger tasks, be more ambitous. Batch instead of back and forth.
Their wordsAutomation/schedules/wake-up/heartbeat are killer feature. Didn't realize it, but missed it dearly once it was off.
Our reading
Stop managing an agent's context by hand: compaction works very well, so keep one thread per task.
Their wordsEvery thought you spend wondering about "is it using 300k or 380k tokens?" is wasted time. Compaction works incredibly well. Don't manage context by hand. One thread = one task.
Their wordsDon't let agents write for humans. Except: debugging steps, inicident reports, some docs. Mechanical stuff where human on other end doesn't matter that much.
Their wordsReview only code with high blast radius. There are <5 central files -- those I review. Database migrations, yes. Closure vs. class, a neater type, variable names, even duplication: I don't care anymore.
Keep reading
The other pieces are the ones whose claims are worded most like this one's. That measures language, not agreement: they may argue the opposite.