Didn’t Google just blog about how they solve this today? cloud.google.com/blog/topics… Secrets are kept entirely out of the agent sandbox. A security proxy injects credentials when needed. The “secure proxy” route to secure agent infra and secrets seems to be a popular pattern.

Nope, both were US bound. Gate K was direct flight to SF, which I ended up missing. Got rebooked on the Seattle flight, and had to transfer to Gate M. I have TSA Pre/Global Entry in the U.S., so this is never an issue for domestic flights.

Missed my direct flight to SF as the incoming flight was delayed, so Delta rebooked me via Seattle, and had to change gates again.

The real challenge is to make sure I cite all my sources correctly. My config has been evolving over months, and I haven’t been meticulous about tracking where most of the ideas I borrowed came from. Making it public without proper attribution doesn’t sit right with me.

It means I look at every line of code. Often this involves just a perfunctory glance, but I’m more concerned about the general shape of things, as it were, as opposed to obsessing over the minutiae (like variable names).

The reason I still review every line of code despite encoding all the software engineering best practices in my agentic workflow, is because sometimes it’s the *exceptions* to these generalized rules that work best. The agent isn’t discerning enough to make these calls (as yet)

At Paris CDG, Terminal 2E, security check: Gate K: no need to take your electronics out Gate M: take all your electronics out and put it in a tray Me: I wasn’t asked to do it in Gate K Security: Gate K is Gate K, Gate M is Gate M Never seen this in any other airport. 😕

Well then. I guess I have to read this for real, haven’t I?

Speaking of metastable failures, one of the longest open tabs in my browser is this. Yet to read the post, but wonder how much of this methodology overlaps with what you did for Aurora. Only way to find out is to get around to reading the literature. arxiv.org/pdf/2606.00942

I found this unpublished blog post from 6y ago (look at the hand drawn diagram lol). I’d totally forgotten about this, but the key ideas here (embracing non-determinism as a first class citizen via “error kernel” centric design etc) work super well as guidelines for my agents.

The change in question came from a refactor. I have a separate “parity test” suite for any and all refactoring, and those have proven very good at preventing bad regressions, both for refactors involving code reorganization, code modifications, and code deletions.

There’s some overlap, but I’ve also been using new ideas for testing code not covered in the testing series, like getting agents to deliberate plant bugs at every layer and see how tests behave, mutation testing etc; which are new (to me) techniques I haven’t covered before.

Integration tests are great, but they don’t replace unit tests; they come later in the process, whereas bugs that can be caught by a good unit test should be. It allows for speedier iteration. As for the guidelines, I’ll try to make my Claude config public, including testing.

We’ll still need end to end tests, and frankly, we’re going to need all the forms of testing I’ve been talking about for a decade. But “unit tests are now slop” doesn’t have to be the foregone conclusion. Unit tests can be very useful even if a human will never again read them.

5 months and many experiments later, this was totally a case of “skill issue” on my part 😅 Like with code, agents don’t magically write great tests If your unit tests are slop (as mine were), that’s because you don’t have proper testing guidelines for your agents (as I didn’t)

Truly, she’s one of the vilest. Just astonishingly and unfailingly awful. She admits she gets a torrent of racist abuse, but engages in it herself. Despicable piece of shit. She and Indian_Bronson, both cut from the same cloth.

I don’t know if I’m the best person to answer this. Build new stuff. There’s still a world of software that’s yet to be built. Start with any tool you use, and see how you can make it better. Ideas are cheap to try out. The more you do, the better you get at picking better ideas.

lol, I refuse to pay for this app, so typos shall remain. It’s also how you know I’m rawdogging every tweet myself, rather than get AI to write it (which I learnt is a thing, esp with ultra long tweets some influencers and execs post)

Unfinished coding side projects are now way easier to complete than unfinished blog posts. I’m perfectly OK with AI writing my code, even the docs. But I flinch at the idea of AI writing my blog posts, which feels like an egregious violation of a fundamental ethnical contract.