@zeeg @mattzcarey I'm a happy paying Sentry customer for 2 businesses, just starting a new project :)
English
John Rodkey
5.9K posts
John Rodkey
@JohnRodkey
CTO/CIO https://t.co/Znq2TdnqnT, Co-founder @nowims, Consultant to @NFL, ex-CIO @Vroomcars (Nasdaq: $VRM), ex-CIO @TexasDirectAuto, Husband & Father.
Houston, TX Katılım Ekim 2012
904 Takip Edilen522 Takipçiler
@JohnRodkey @mattzcarey we have a startups program but it requires certain requirements - otherwise we'd just be giving away the business to anyone who asked
right now its limited to certain programs (e.g. YC) but i know we're slowly expanding it
English
@mitchellh Agreed. We build the disclosure into the opening statement from the ai agent. For example, "Hello, I'm Mitchell an ai receptionist for xyz corp..." then we allow the user to get out of the ai flow if they want by asking for a human.
English
It's so insanely disrespectful for an AI agent to talk to real people without consent or at least disclosure. This is the type of stuff I'm hugely supportive of government regulation. The FCC must expand the definition of robocalling and TCPA-style regulation to online AI.
English
@matteocollina What are you using on mobile? And what is your mobile device?
English
Most people are missing out on mobile coding.
@levelsio@levelsio
Are you guys aware I am coding mostly on my phone now all day via Termius to Claude Code on my server while I go with gf to the dentist, clothing store, cafe, etc. 😛✌️
English
@DavidKPiano That’s just crazy, not doing code review. I treat ai like any other dev, I want somebody performing a code review to ensure we don’t overlook anything and considered all possible scenarios.
English
Not reviewing the code is just laziness
Not talking about text/styles or reading every single token; any code relevant to the main critical application logic should be reviewed
Good engineers can pattern-match large blocks of code and OK it in a few seconds. Don't be lazy.
andrew chen@andrewchen
One question I've been asking founders is: do you try to review all the code that the LLMs write or do you just accept it? I think it's about 50-50 right now but the momentum is towards just accepting the AI-generated code and I think that number will eventually go to 100% This is one of the most telling indications of how AI-native a team is. It's hard to get super high throughput if you are reviewing every line Poll: what do you do?
English
Everyone talking about slop forks, i’ve always eating my slop with a spoon… more surface area
English
@floydophone it just did a fine job of cleaning up screenshots and putting them in folders for me by content and dates
English
@bcherny do you still recommend claude.md? also, this page need to be updated now that you have a 1M context window code.claude.com/docs/en/best-p…
English
🎁 Happy Friday - Opus 4.6 1M is now the default Opus model for Claude Code users on Max, Team, and Enterprise plans.
Pro and Sonnet users can opt in with /extra-usage.
Claude@claudeai
1 million context window: Now generally available for Claude Opus 4.6 and Claude Sonnet 4.6.
English
@BenjDicken The only thing missing is a networking book. Swap in Kurose & Ross and you've got the entire OSI model covered
English
@JohnRodkey @gdb code review is right there in the desktop app. I use it almost exclusively, but occasionally browse to GitHub to see the full diff in a multi-commit PR
English
I read the OP. The AI deleted an environment because it doesn't understand "blast radius"—which is exactly what architecture "the pieces" is. And Amazon's literal fix is making seniors review junior/AI code. Which is exactly what I just said. As a former CIO of a $1B corp, I would’ve done this from the start. One day I think AI will understand the pieces better, but not today.
English
@JohnRodkey @RyanCarniato If you're still assuming "AI will just write the code" then you're not paying attention. Look at the OP ffs
English
Something nobody’s talking about:
Senior engineers have to review AI code. Cool. But how do junior and mid-levels become senior without years of writing code by hand?
Smashing the magic button and tossing the result to a senior engineer for review does not make you a senior
Lukasz Olejnik@lukOlejnik
Amazon is holding a mandatory meeting about AI breaking its systems. The official framing is "part of normal business." The briefing note describes a trend of incidents with "high blast radius" caused by "Gen-AI assisted changes" for which "best practices and safeguards are not yet fully established." Translation to human language: we gave AI to engineers and things keep breaking? The response for now? Junior and mid-level engineers can no longer push AI-assisted code without a senior signing off. AWS spent 13 hours recovering after its own AI coding tool, asked to make some changes, decided instead to delete and recreate the environment (the software equivalent of fixing a leaky tap by knocking down the wall). Amazon called that an "extremely limited event" (the affected tool served customers in mainland China).
English
@RyanCarniato @AdamRackis I think writing code is the easy part. AI handles that. Juniors need to focus on breaking the pieces down—the architecture. My solution: Juniors write prompts, then seniors review the AI code and explain the architecture to level them up. What are y’all seeing?
English
@AdamRackis I get that. You don't learn without practical application. What jumping in looks like is changing. Maybe at a rate no one is equipped currently with how to deal with. But it isn't unusual to find this separation in other engineering fields.
English
@samlambert @PlanetScale i prefer Zero or whats directly above it
English
@LyalinDotCom @geminicli i have been using the cli more and more with 3.1 and it is fantastic so far. great work to you and the team!
English
We're actively working on making @geminicli startup time faster across MacOS and Windows. We know we have to do better here and it's coming.
English
it won't stop it won't freaking stop every day another LLM telling you that the code your LLM generated is shit
dax@thdxr
ok it's been a month and it's even worse it's like a war zone rn
English
@cramforce Opus “gets me” faster but 5.3 codex will one shot anything well specced and just never stop until it’s done. I’m still not 100% sure about 5.4 because it kept stopping and saying “I’ll do this next.” And then… not doing it.
English
lmao
a16z@a16z
"Not having a coding experience is becoming an advantage." Replit CEO Amjad Masad: "You don't need any development experience. You need grit. You need to be a fast learner." "If you're a good gamer, if you can jump in a game and figure it out really quickly, you're really good at this." "Coders get lost in the details." "Product people, people who are focused on solving a problem, on making money, they're going to be focused on marketing, they're going to be focused on user interface, they're going to be focused on all the right things." "I think this year it's gonna flip, and I think not having a coding background is gonna be more advantageous for the entrepreneur." @amasad with @jackhneel
HT
@kentcdodds Which model do you prefer or do you switch based on domain/feature?
English
As someone who's already at the spot where agents write 100% of my code, it's crazy to me to see that 28% of responses here think they still won't be there a year from now
Ryan Florence@ryanflorence
By next year I suspect agents will write almost all (or actually all) of my code
English