HackerTrans
トップ新着トレンドコメント過去質問紹介求人

samber

no profile record

投稿

Trump administration asks OpenAI to stagger release of GPT 5.6

old.reddit.com
1 ポイント·投稿者 samber·18 日前·0 コメント

Anthropic Policy on the AI Exponential

anthropic.com
2 ポイント·投稿者 samber·先月·1 コメント

Linear Agent

linear.app
3 ポイント·投稿者 samber·先月·0 コメント

SemiAnalysis: TCO of Space Datacenters

newsletter.semianalysis.com
4 ポイント·投稿者 samber·先月·0 コメント

IEA: About Energy and AI

iea.org
3 ポイント·投稿者 samber·先月·0 コメント

The engine behind the 100 TB GitHub search engine

github.blog
8 ポイント·投稿者 samber·2 か月前·0 コメント

UX Research Tooling Landscape

github.com
2 ポイント·投稿者 samber·3 か月前·0 コメント

A collection of 35 Golang Agent Skills

github.com
3 ポイント·投稿者 samber·4 か月前·2 コメント

Go code is leaving 90% of the CPU idle until now

samuelberthe.substack.com
1 ポイント·投稿者 samber·4 か月前·0 コメント

Claude-swarm-monitor: track progress of your Claude Code agents

github.com
2 ポイント·投稿者 samber·5 か月前·0 コメント

Show HN: A web-based visualizer for Go module dependency graph

go-mod-graph.samber.dev
1 ポイント·投稿者 samber·6 か月前·0 コメント

Create illustrations for your GitHub project

samuelberthe.substack.com
1 ポイント·投稿者 samber·9 か月前·0 コメント

Go beyond Goroutines: introducing the Reactive paradigm

samuelberthe.substack.com
68 ポイント·投稿者 samber·9 か月前·41 コメント

Samber/lo v1.52.0 – now supports Go 1.23's iterators

github.com
1 ポイント·投稿者 samber·9 か月前·0 コメント

Why Your 'Optimized' Code Is Still Slow: Faster Time Comparison

samuelberthe.substack.com
1 ポイント·投稿者 samber·10 か月前·0 コメント

Critical TTL patterns for in-memory caching

samuelberthe.substack.com
3 ポイント·投稿者 samber·10 か月前·0 コメント

Vibe-coding and open-source: 286k LoC, 2 months

github.com
4 ポイント·投稿者 samber·10 か月前·0 コメント

Moe Inference Economics from First Principles

tensoreconomics.com
2 ポイント·投稿者 samber·10 か月前·0 コメント

Show HN: The Great GPT Firewall

github.com
1 ポイント·投稿者 samber·11 か月前·0 コメント

Spurious correlations (correlation is not causation)

tylervigen.com
3 ポイント·投稿者 samber·11 か月前·0 コメント

コメント

samber
·28 日前·議論
Comparing compute cost versus FlashAttention-2 is not very honest to me.

FlashAttention-2 is not used anymore for at least 2y.

This architecture would have been a massive improvement 3 years ago, but it is a ~solved~ problem IMO.
samber
·28 日前·議論
According to Subquadratic, Needle in a Haystack is strong up to 12m tokens, but RULER has not been tested above 128k tokens ??
samber
·2 か月前·議論
OMG. I'm going to recode some of my libraries.
samber
·4 か月前·議論
35+ atomic skills covering all aspects of the language (conventions, common errors, top libraries, testing, benchmarks, performance, troubleshooting, etc.).

Benchmarks I ran on Opus 4.6 show a 43% reduction in Go errors and bad practices.

Install with: npx skills add -g https://github.com/samber/cc-skills-golang --skill '*'
samber
·9 か月前·議論
IMO, this is much more readable.

So many Go developers ignore some tools because they consider them "not idiomatic".

But why not use abstractions when available ??? Did we forget to be productive ?
samber
·10 か月前·議論
`do` builder here

After 2y in beta, I’ve just released v2 of “do”, the dependency injection toolkit for Golang.

This major version introduces a new scope-based architecture, transient services, interface binding, improved dependency tracking, and circular dependency detection.

Error handling and service naming are more consistent, and based on your feedback, a troubleshooting UI has been added.

A new LLM-ready documentation is available, featuring numerous demos you can run in 1 click: https://do.samber.dev/

Read the full changelog here: https://github.com/samber/do/releases/tag/v2.0.0

Migration from v1: https://do.samber.dev/docs/upgrading/from-v1-x-to-v2
samber
·昨年·議論
I wrote my own error wrapper: github.com/samber/oops

Example:

err := oops. Code("iam_missing_permission"). In("authz"). Tags("authz"). Time(time.Now()). With("user_id", 1234). With("permission", "post.create"). Hint("Runbook: https://doc.acme.org/doc/abcd.md"). User("user-123", "firstname", "john", "lastname", "doe"). Errorf("permission denied")

For easier debugging, the error contains the full stacktrace.
samber
·昨年·議論
Elasticsearch is a good bet, if you need to use multiple filters with your queries, and when you grow above the acceptable size of an in-memory database.