Bruno

Bruno

@lucattelli.bsky.social

software engineer

104 followers70 following281 posts

I only read truths :)

bethany@trustyduck.dev

Corporate Slack SEO tips: - send a message during or right after a meeting where you know your audience is in, they’ll likely be checking slack before working on something else - rage bait by proposing a potentially wrong answer to your question - don’t @here, it will just annoy people

Bruno reposted

Good writing is writing that doesn't sound like everyone else. Which also means good writing doesn't sound like it was written by AI. So the way students can avoid getting accused of using AI to write their essays is to develop a strong and unique authorial voice.

aaronrosspowell.comThe Answer to Student AI Writing Slop is Authorial Voice - Aaron Ross PowellLLMs are bad at sounding unique. So we should help young writers to develop their own unique prose style.

Bruno reposted

Bruno reposted

Anthropic engineering abandons single global merge queue, suddenly discovering that singular global bottlenecks make scaling hard They’re replacing it with the least obvious solution — merge untested code and hope it’s caught pre-deploy www.anthropic.com/engineering/...

Got my 10th gen i7 Legion 5i back recently. Removed Windows and installed Debian 13 this time (not Arch). Claude was really helpful in helping me troubleshoot configuration issues. It was also super helpful in writing documentation about my process. I now have a repo with all steps!

Software fundamentals matter more as AI scales: concurrency, local state, message passing, cancellation. This paper applies them to LLM workers: a 0.6B model solves 72% of 25×25 Sudoku puzzles; a larger model roughly halves document QA latency with similar accuracy. arxiv.org/abs/2607.01077

arxiv.orgMessage Passing Enables Efficient ReasoningWhile inference-time scaling has improved the reasoning abilities of large language models (LLMs), the need to generate long chains-of-thought (CoTs) is a computational bottleneck. Thus, in contrast t...

Andrew Ng's team went through 10k+ job posts and boiled AI engineering down to 4 skills: shipping AI apps, solid software fundamentals, knowing how to drive coding agents, and having enough product sense to decide what to build. www.deeplearning.ai/the-batch/th...

deeplearning.aiAndrew Ng and DeepLearning.AI Present the Essential Skills Every AI Engineer Needs in 2026I am delighted to present The AI Engineering Skills Map. AI allows us to build software very differently today than in 2022, and everyone with...
View full thread

the workflow is the old one (plan → build/verify → deploy/monitor), just with less time on code and more on specs and review. Nice callout: the "let it run for hours on millions of tokens" hype is overrated. Tight iteration wins. www.deeplearning.ai/the-batch/th...

deeplearning.aiHow to Use Coding Agents Effectively from Planning to Execution and MonitoringA key AI engineering skill is using coding agents. Your skill at steering them both to write code and to carry out non-code tasks, such as...

Andrew Ng's team went through 10k+ job posts and boiled AI engineering down to 4 skills: shipping AI apps, solid software fundamentals, knowing how to drive coding agents, and having enough product sense to decide what to build. www.deeplearning.ai/the-batch/th...

deeplearning.aiAndrew Ng and DeepLearning.AI Present the Essential Skills Every AI Engineer Needs in 2026I am delighted to present The AI Engineering Skills Map. AI allows us to build software very differently today than in 2022, and everyone with...