San Francisco, CA
- Tweets 11,744
- Following 165
- Followers 47,177
- Likes 35,673
People talk about the diminishing returns of writing unit tests in the agentic era, but what’s even less useful? Comments.
As a rule, drop all comments. If something is important enough to need a comment, make an exception for it and audit the comment so it actually makes sense.
40
3
112
12,872
I need a personal agent for work where it:
- triages all incoming bugs
- manages slack comms
- gives me a summary and classification of all emails (8000+ unread), design docs, SEVs etc
I don’t need an agent for my personal life, where I just don’t deal with anything like this.
14
109
9,028
Here’s an example where Twitter Engineering was way ahead of the curve.
Many companies didn’t have to deal with “Twitter scale”, but agentic architectures are fast changing the scale even small orgs are expected to handle re stateful load balancing, orchestration, storage etc.
This post on the evolution of client side (distributed w/o coordination) load balancing at Twitter is 🤯
- [service] "mesh" topology + P2C has cons
- the "random aperture" approach which scales unfairly
- "deterministic random aperture" scalable + fair
bit.ly/2FvTH2u
7
10
194
18,700
Ball-knowers knew all along that Twitter wasn’t just going to collapse after Elon fired 75% of the staff, because Twitter 1.0 had some of the best infrastructure engineers working on it, and when you get that calibre of people building your foundations, it lasts.
18
7
390
44,762
I agree! But an awful lot of self-proclaimed experts, who seemed to have the right credentials, predicted various dooms that didn't happen! Certainly it means Twitter's eng was way more robust than people thought. But these are statements about the experts, not the engineering.
1
5
394
TBH, there was a lot of hyperbole coming out of all sides in Nov 2022.
Some of us tried to inject some sanity into the discussion and said that we might not see the fail whale but that’s *because* of the work done over a decade, but our voices got drowned out in the cacophony.
Twitter going back to 1K employees might not cause the fail whale, but that’s *because* of the work done by 1000s of the engineers who got unceremoniously fired in the past 3 weeks.
The VC and CEO class are going to come away with all the wrong takeaways from this Twitter saga.
3
511
To be a fair, a lot of *tech* people made this prediction. I remember people who *worked at Twitter* saying things like "I give it two weeks, they just fired the crucial person who keeps X up, once that goes down nobody will be able to restart it and you'll shutdowns propagate".
1
17
1,616
There's been a ton of talk about the role of humans in code review, and when and how humans should be signing off on changes.
I believe that, long-term, humans have no role in routinely reviewing code.
137
166
1,711
506,512
I'm quite hopeful for what we have cooking at Antithesis. Perfect? Nope. But we have a hell of a verification to build around. Still plenty of issues to work out though.
But also, this general shape of things has been my bag for a long long time, so I'm "Captain Priors in the House" over here.
1
5
3,671
The bigger problem is that people often open PRs for bugs etc that don’t fix the problem at all.
LLM often manage to even repro something, but it’s for a really obscure edge case no real end user is hitting, while the actual bug goes undiagnosed.
3
24
5,695
With agentic coding, even in the best case scenarios (where you read every line of generated code), there’s something we’ve lost, perhaps irrevocably.
That has a real cost, & not to mention, second and third order effects, that “software factories” etc don’t necessarily solve.
24
13
180
7,866
I saw this happen on a Reddit forum for SF last week when someone posted an AI-generated event page. The person got so much hate that they canceled the event. A lot of people suggested Canva as if that is not AI 🫠
I talked to a coffee shop owner in Buffalo who was getting destroyed on Instagram after showing a photo of a fall menu poster she made with ChatGPT. One person with a big local following was DMing her and threatening to post about it she woudln't agree to hire an artist to redo it. But it turned out it was more complicated that her skipping out on hiring someone: she told me she's never hired anyone to make their chalkboard menu signs - either she did posters herself with photos and Canva or a staffer did in chalk. The AI menu was just her trying to reduce her own workload (she has a newborn and wanted to save time) businessinsider.com/coffee-s…
1
6
2,014
Writing a post about “how to effectively debug code when you no longer fully understand it”.
This is real bug I had to fix in ~9K LOC (100% AI generated).
None of the tactics I started with here helped “root cause” the issue, lol. Had to take a different approach to isolate it.
33
23
1,038
52,188
One thing I don’t see many talk about is how coding agents can often *amplify* weaknesses, both individual and organizational.
I think of this as the “empty calories” problem, which is exacerbated when “hyperproductivity” originates from your team’s “lowest common denominator”.
13
24
231
12,880
I used to say “you should aim to understand every line of generated code” (and I still try to!)
But in practice, there’s an emergent set of (sometimes counterintuitive) skills (NOT loops/toxenmaxxing!) that lets one ship and debug large amounts of unfamiliar code productively.
5
2
94
17,017
About 40% done. Hoping to fully wrap this up next week.
Posting here to keep myself accountable, and not abandon this draft, lol.
2
10
1,333
Update: Been super busy lately, but should have time on weekends in Sept.
Should be done by the end of this month, I hope. Here’s an excerpt (as proof I haven’t abandoned this, lol)
One thing I don’t see many talk about is how coding agents can often *amplify* weaknesses, both individual and organizational.
I think of this as the “empty calories” problem, which is exacerbated when “hyperproductivity” originates from your team’s “lowest common denominator”.
2
6
1,795
The IDE/editor might be dead, but it helps to occasionally open one and read some code.
Often, what “looked OK” in isolation when an agent wrote a patch can end up looking *very* different from more of a “panoramic view”, with misshapen and incoherent APIs, abstractions.
10
3
51
4,131
I previously called this an “emergent” skill set.
The people yapping about how you “don’t need to understand the code anymore” are NOT telling you how to sharpen or even hone this new skillset, where you *can* debug problems out of your comfort zone, methodically with AI.
I used to say “you should aim to understand every line of generated code” (and I still try to!)
But in practice, there’s an emergent set of (sometimes counterintuitive) skills (NOT loops/toxenmaxxing!) that lets one ship and debug large amounts of unfamiliar code productively.
2
6
1,894
So the emergent skill is still human. The AI made the cheap checks affordable. No LLM did the abductive reasoning; the hypothesis formation and skepticism were all yours.
1
61
Incidentally, this proves that with AI you can now solve problems *way* out of your wheelhouse or your “knowledge/skill” level, if you persevere enough and can get creative with how you approach debugging.
It’s a different skill set than coding or even traditional debugging tho.
The people who blithely claim “you don’t need to understand the code anymore” aren’t telling you what debugging looks like in such cases
Here’s one example where I had to debug a gnarly problem at work with Claude *without* access to the code
Possible? Yes
“Easy”? Nope 🙂↔️
5
2
75
9,437