Idk, do you rly think Bernie’s not going to have much influence in the 2028 Dem primary?
Or that any Dem primary candidate can call out the virulent anti-Semitism seen in some pockets of the far-left, without getting dragged for it, and being able to brush it off?
Doubtful.
Or that any Dem primary candidate can call out the virulent anti-Semitism seen in some pockets of the far-left, without getting dragged for it, and being able to brush it off?
Doubtful.
1
0
0
46
The emboldened left isn’t going anywhere. I want to see a non-identitarian and non-slopulist Dem candidate win the primary, but most will end up pandering to various activist groups on all manner of issues. It’s going to be utterly insufferable.
1
1
6
1,216
About half my tests are set up just to test the non-determinism in the services.
I find keeping the non-determinism isolated to specific parts of the code helps agents to then test the hell out of just those parts.
If those invariants hold true, I trust other parts to work.
I find keeping the non-determinism isolated to specific parts of the code helps agents to then test the hell out of just those parts.
If those invariants hold true, I trust other parts to work.
0
0
3
506
CI as it currently exists (build all the code to run the full test suite) seems like a waste of compute, which can be better assigned to agents to analyze and test select parts of code but way more rigorously.
My pre-merge tests don’t do full builds; just targeted micro-tests.
My pre-merge tests don’t do full builds; just targeted micro-tests.
How do you deal with the CI time to run all those tests, etc?
20
6
149
8,957
Idk, I think the Democratic base really wants to see accountability (rightfully so). For Jan 6 to begin with. But also the mind-boggling amount of corruption and self-dealing.
The backlash against the Dem establishment will be harsh if they’re seen as cutting deals.
The backlash against the Dem establishment will be harsh if they’re seen as cutting deals.
1
1
14
2,096
A mix of the techniques shown here, in short.
This has been my dream for nearly a decade now, and with agents I’m finally able to see this (partially) come true.
But turns out this is what I necessarily require to trust agent generated code. It’s not optional anymore, imo.
This has been my dream for nearly a decade now, and with agents I’m finally able to see this (partially) come true.
But turns out this is what I necessarily require to trust agent generated code. It’s not optional anymore, imo.
21
18
345
23,187
Revisiting my tweets from 6m ago is funny, because I don’t run into any of these problems now at all.
What was missing in my setup 6 months ago was the verification process I now use.
70% of my daily token spend goes into verifying patches, and only ~20-30% into creating them.
What was missing in my setup 6 months ago was the verification process I now use.
70% of my daily token spend goes into verifying patches, and only ~20-30% into creating them.
Might be anecdotal, but just this week multiple AI generated PRs with subtle bugs got merged that required several additional days and a lot of manual verification to fix.
“Velocity” isn’t going up when you consider how much effort was spent fixing things post-merge, pre-deploy.
23
8
223
25,450
This is 100% true.
Your reliability in general should be *higher* with agent written code, as measured by error rates, SLO violations, client regressions etc.
If you’re having more or equal incidents now than when you were manually writing code, that’s likely a “skill issue”.
Your reliability in general should be *higher* with agent written code, as measured by error rates, SLO violations, client regressions etc.
If you’re having more or equal incidents now than when you were manually writing code, that’s likely a “skill issue”.
Agreed. We have shockingly few production errors these days. E.g. panic from a missing nil-check happens once every couple of months, and usually involves legacy code humans partially wrote. At this point, if I write it by hand you are going to get more bugs.
16
17
317
15,286
Mostly timing related things, and benign things related to ordering of some things.
The good thing is an agent can be quite discerning about these things, provided you give it the general boundaries of correctness.
The good thing is an agent can be quite discerning about these things, provided you give it the general boundaries of correctness.
0
0
3
328
Very important. I first started using the idea of “planted bugs” mainly to ensure unit tests weren’t slop (because a lot of agent written tests passed even when the code itself broke), but now I love it for refactors, reliability tests (with broken clients), and more.
0
0
1
230
How to design good verification loops?
Well, that’s most of the job now, isn’t it?
Last week, I finally landed a refactor I started in Jan. It’s good I waited b/c the Jan version was buggy af.
This is the verification process I used for refactors (have different ones for perf)
Well, that’s most of the job now, isn’t it?
Last week, I finally landed a refactor I started in Jan. It’s good I waited b/c the Jan version was buggy af.
This is the verification process I used for refactors (have different ones for perf)
How does one ensure verification loops are actually correct? Rigorous needs to take cost into consideration as well.
26
16
522
24,136
This won’t make it an “expert” with 10 years of experience.
But it *will* produce better code than it’ll do without any of the guidelines. And most often, that’s “good enough”.
But it *will* produce better code than it’ll do without any of the guidelines. And most often, that’s “good enough”.
1
0
2
690
You missed the “domain references” part. This includes
- watching all conf videos by experts
- reading code in popular projects using the domain
- reading commits/tweets/blogs from the known experts
- learn all domain specific perf/test tooling
- build tools for missing features
- watching all conf videos by experts
- reading code in popular projects using the domain
- reading commits/tweets/blogs from the known experts
- learn all domain specific perf/test tooling
- build tools for missing features
3
0
16
1,124
Getting agents to produce “high-quality, optimized outputs in a language or domain you’re personally unfamiliar with” is an emergent skill.
It doesn’t happen magically, but requires:
1. A change in mindset
2. Setting up the best domain references and rigorous verification loops
It doesn’t happen magically, but requires:
1. A change in mindset
2. Setting up the best domain references and rigorous verification loops
I still can't believe people are rewriting software into Rust for performance reasons without learning any Rust.
I'm optimizing a Go program right now and using Agents to do the work, but I have to steer them a lot or they get stuck in local minima, but I can only steer them because I have 10 years of doing performance work in Go.
If I was optimizing a Rust program I'd have to just keep saying: "make it faster" over and over again until it got stuck somewhere dumb.
10
5
238
11,984
Returning from international travel, followed by jet lag, physical exhaustion, remnants of a throat infection I picked up when traveling, dealing with a case of misplaced luggage across 3 carriers, catching up on work on a quasi-zombie state …
Yeah, I’m glad its the weekend ❤️
Yeah, I’m glad its the weekend ❤️
0
0
21
3,783
IMO it’s easy to mistake “AI positivity” for general “positivity”, which this app most certainly doesn’t promote.
Even “AI positivity” can be an echo chamber representative of the frontier labs and those at the bleeding edge. No way representative of the broader tech industry.
Even “AI positivity” can be an echo chamber representative of the frontier labs and those at the bleeding edge. No way representative of the broader tech industry.
2
1
26
1,996
Didn’t Google just blog about how they solve this today? cloud.google.com/blog/topics/ai…
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.
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.
Is anyone solving how we give secrets/passwords to agents in a way that doesn't expose them to the upstream provider? It seems like codemode could be used to pass the tokens around and you'd only need to approve it for where the token is actually consumed
23
25
468
29,104