Stéphane Petit

Random Software Developer. Code lover, caring about Craftsmanship, Empathy and Ethics.

[Stéphane Petit @spetit95]

Never forget again to wish the birthdays of your friends with birthday-pal https://t.co/vyZCosIMRS - A CLI tool written in #GoLang #sideproject

[Camille Roux @CamilleRoux]

🔎 claude-real-video permet à un Claude de vraiment "regarder" une vidéo : extraction de frames par détection de changement de scène, déduplication, transcription audio. Fonctionne en local, MIT, depuis une URL ou un fichier. ⬇️

[Lenny Rachitsky @lennysan]

"Using coding agents well is taking every inch of my 25 years of experience as a software engineer, and it is mentally exhausting.

I can fire up four agents in parallel and have them work on four different problems, and by 11am I am wiped out for the day.

There is a limit on human cognition. Even if you're not reviewing everything they're doing, how much you can hold in your head at one time. There's a sort of personal skill that we have to learn, which is finding our new limits. What is a responsible way for us to not burn out, and for us to use the time that we have?" @simonw

[Boris Cherny @bcherny]

Mistakes happen. As a team, the important thing is to recognize it’s never an individuals’s fault — it’s the process, the culture, or the infra.

In this case, there was a manual deploy step that should have been better automated. Our team has made a few improvements to the automation for next time, a couple more on the way.

[antirez @antirez]

Biggest mistake in AI coding era: to believe that specifications should be either natural language OR something else. The best combo is a natural language high level specification (the intend), plus code (as it gets written) documenting the finer behaviors.

[Graeme Rocher @graemerocher]

Zero surprises here. AI is an incredible tool in the hands of experienced engineers with domain knowledge sufficient to drive the AI to do the right thing. For this you need a clear understanding of application architecture from end to end.

[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).

[Stéphane Petit @spetit95]

@CamilleRoux Il y a beaucoup d’articles de ce genre en ce moment mais il ne faut pas les prendre pour acquis, souvent les raisons sont bien plus nuancées que cela:

https://t.co/8gpV8nxPSc

https://t.co/sczV4d5zE7

[Jarred Sumner @jarredsumner]

@om_patel5 - LSP is enabled by default
- Sadly, it doesn’t make Claude 600x faster
- it is kind of popula but probably too hard to use. It works well in some codebases.

[Harrison Kinsley @Sentdex]

I remember trying to force myself to use C++ to do the same things I could do 100x faster in Python. I just couldn't do it. Not because it was hard, because there was no point for that particular problem. There are obviously still many important and useful usecases for C++ today, but the needs have changed over time.

Abstraction is a good and natural thing. It's a core human capability that has given us our entire civilization and allows us to progress technologically.

When assembly hit, the machine coders disliked it. When C++ came around, the C/Assembly people disliked it. C++ people still often dislike Python. It's definitely okay to be sad about change, and Mo's take here seems super human and genuine, recognizing mostly that times are changing, but the argument is false.

Mo is not useless. Mo just needs harder problems to solve.

[Thariq @trq212]

Today we're launching local scheduled tasks in Claude Code desktop.

Create a schedule for tasks that you want to run regularly. They'll run as long as your computer is awake.

[Camille Roux @CamilleRoux]

J'ai vibe codé un assistant de recherche d'appart avec Claude Code : une slash command qui scrape 3 sites en browser automation, score les annonces selon mes critères, contacte les agences.
Dev en quelques heures, 30 min/jour gagnées (et beaucoup de charge mentale en moins) !

[Camille Roux @CamilleRoux]

Le MCP est mort, vive le CLI : les LLM n'ont pas besoin d'un protocole spécial, ils utilisent déjà très bien les outils en ligne de commande
https://t.co/7M3GrnFh7c

[Stéphane Petit @spetit95]

@_smontlouis Et en vrai, ta problématique risque d’arriver aussi avec les équipes PO/QA. Comment vont-ils gérer et tester tous ces changements qui arriveront plus nombreux et plus vite. La réorganisation des équipes IT ne fait que commencer 😅

[Stéphane Petit @spetit95]

@_smontlouis Avant d’arriver à la PR,comment tu fais toi pour relire et contrôler le code généré par IA ? Au delà d’une certaine quantité de code, on a tendance à accepter tous les changements en mode YOLO et on arrive au même souci. AMHA,le découpage des US semble + que jamais indispensable

[Stéphane Petit @spetit95]

@_smontlouis Quelques trucs basiques, du tri dans mes fichiers et du ménage, c’est toujours bon à prendre. Il m’a aussi pondu un excel pas mal fichu pour gérer mes comptes car j’ai des use cases un peu spécifiques, et il s’est plutôt bien débrouillé

[Addy Osmani @addyosmani]

Boris created Claude Code. His point here is important - when AI handles the code generation, the engineer's value shifts to the decisions above the code:

1. what do we build?
2. why? for whom?
3. and how it all fits together.

The bottleneck was always judgment, taste, and systems thinking. AI just made that more obvious.

[Camille Roux @CamilleRoux]

📢 Développeur à l’ère des agents IA : mutation, industrialisation et avenir du métier #veille
https://t.co/IvinrSCRKS

[Jaana Dogan ヤナ ドガン @rakyll]

I'm not joking and this isn't funny. We have been trying to build distributed agent orchestrators at Google since last year. There are various options, not everyone is aligned... I gave Claude Code a description of the problem, it generated what we built last year in an hour.

[Camille Roux @CamilleRoux]

📢 Les bons et les mauvais devs #veille
https://t.co/5ccbUIJ0sL

[Rom1 @Rom1CHAUSSEDOUX]

Le mirage du Vibe Coder Promettre la fin des devs : cas d'école de l'effet Dunning-Kruger assisté par IA, c'est confondre facilité de production avec maîtrise de l'ingénierie. Voici pourquoi ce conseil est dangereux pour votre carrière (et pour l'infrastructure mondiale). ⬇️

[Ben Dicken @BenjDicken]

ORMs are great, but you still need to learn SQL + database performance fundamentals.

Running an app at scale demands a deeper understanding of indexing, joins, and data access patterns.