什么火了

今日热点 · 前 3

一件事为什么火、在哪些平台上榜、热度是涨是落。

看今日热点 →

AI 热点 · 前 3

只看和 AI、科技有关的热点。

看 AI 热点 →

最近的重大发布

各家 AI 模型的发布、开源和传言,官方消息和媒体报道分开标。

打开时间线 →

综合 · 热门站点

全部综合站点 →

科技 · 热门站点

全部科技站点 →

娱乐 · 热门站点

全部娱乐站点 →

游戏 · 热门站点

全部游戏站点 →

生活 · 热门站点

全部生活站点 →

知识 · 热门站点

全部知识站点 →

社区 · 热门站点

全部社区站点 →

快讯来源

各家快讯一页排开,重要的标出来。

全部快讯来源 →

AI 资讯

常读的 AI 公众号和科技媒体,新文章都在这一页。

去看 AI 资讯 →

按平台

社媒

公众号、Telegram、Newsletter、个人博客,还有外媒和 AI 公司的博客,新内容按时间排好。

看全部社媒 →

品牌

各家今天的金价多少钱一克,从低到高排好。

看今日金价 →

今天最低的 3 家

看足金饰品榜 →
全部/科技/Eugene Yan · Writing

技术博客与 newsletter feed · Eugene Yan · Writing

HISTORY2026年11月21日214 不同热搜 · 统计中
10/23—11/21 有历史数据
DAILY UNIQUE TOPICS214 条热搜今天 00:00 起累计,每次采集后更新
热搜话题最高排名首次上榜离榜时间在榜时长上榜次数
01
AI Engineer 2026 - Using LLMs to Secure Source Code

Sharing what we learned from Project Glasswing.

#100:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
02
Patterns for Building Cybersecurity Evals

A sandboxed target, inputs that influence task difficulty, tools, and a grader.

#200:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
03
Using LLMs to Secure Source Code

Build a threat model, discover vulnerabilities, verify, triage, and patch.

#300:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
04
How to Work and Compound with AI

Context as infra, taste as config, verify for autonomy, scale via delegation, close the loop.

#400:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
05
2025 Year in Review

An eventful year of progress in health and career, while making time for travel and reflection.

#500:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
06
Product Evals in Three Simple Steps

Label some data, align LLM-evaluators, and run the eval harness with each change.

#600:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
07
Advice for New Principal Tech ICs (i.e., Notes to Myself)

Based on what I've learned from role models and mentors in Amazon

#700:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
08
Training an LLM-RecSys Hybrid for Steerable Recs with Semantic IDs

An LLM that can converse in English & item IDs, and make recommendations w/o retrieval or tools.

#800:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
09
Evaluating Long-Context Question & Answer Systems

Evaluation metrics, how to build eval datasets, eval methodology, and a review of several benchmarks.

#900:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
10
AI Engineer 2025 - Improving RecSys & Search with LLM techniques

Recsys & search are converging with LLMs via semantic IDs, data augmentation, and unified foundation models.

#1000:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
11
Exceptional Leadership: Some Qualities, Behaviors, and Styles

What makes a good leader? What do good leaders do? And commando, soldier, and police leadership.

#1100:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
12
Building News Agents for Daily News Recaps with MCP, Q, and tmux

Learning to automate simple agentic workflows with Amazon Q CLI, Anthropic MCP, and tmux.

#1200:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
13
An LLM-as-Judge Won't Save The Product—Fixing Your Process Will

Applying the scientific method, building via eval-driven development, and monitoring AI output.

#1300:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
14
Frequently Asked Questions about My Writing Process

How I started, why I write, who I write for, how I write, and more.

#1400:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
15
NVIDIA GTC 2025 - Building LLM-Powered Applications

Chip Huyen and I share what we've learned, best practices, and insights at NVIDIA GTC 2025.

#1500:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
16
Improving Recommendation Systems & Search in the Age of LLMs

Model architectures, data generation, training paradigms, and unified frameworks inspired by LLMs.

#1600:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
17
Building AI Reading Club: Features & Behind the Scenes

Exploring how an AI-powered reading experience could look like.

#1700:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
18
2024 Year in Review

A peaceful year of steady progress on my craft and health.

#1800:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
19
A Spark of the Anti-AI Backlash (on Bluesky)

How the sharing of 1M Bluesky posts uncovered the strong anti-AI sentiment on Bluesky.

#1900:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
20
Seemingly Paradoxical Rules of Writing

With regard to writing, there are many rules and also no rules at all.

#2000:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
21
How to Run a Weekly Paper Club (and Build a Learning Community)

Benefits of running a weekly paper club, how to start one, and how to read and facilitate papers.

#2100:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
22
My Minimal MacBook Pro Setup Guide

Setting up my new MacBook Pro from scratch

#2200:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
23
39 Lessons on Building ML Systems, Scaling, Execution, and More

ML systems, production & scaling, execution & collaboration, building for users, conference etiquette.

#2300:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
24
AlignEval: Building an App to Make Evals Easy, Fun, and Automated

Look at and label your data, build and evaluate your LLM-evaluator, and optimize it against your labels.

#2400:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
25
Weights & Biases LLM-Evaluator Hackathon - Hackathon Judge

Being a human judge at the Weights & Biases LLM-as-a-Judge Hackathon

#2500:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
26
Building the Same App Using Various Web Frameworks

FastAPI, FastHTML, Next.js, SvelteKit, and thoughts on how coding assistants influence builders' choices.

#2600:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
27
Evaluating the Effectiveness of LLM-Evaluators (aka LLM-as-Judge)

Use cases, techniques, alignment, finetuning, and critiques against LLM-evaluators.

#2700:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
28
How to Interview and Hire ML/AI Engineers

What to interview for, how to structure the phone screen, interview loop, and debrief, and a few tips.

#2800:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
29
AI Engineer 2024 Keynote - What We Learned from a Year of LLMs

Special double-feature closing keynote from the 6 authors of the hit O'Reilly article on Applied LLMs.

#2900:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
30
Netflix PRS 2024 - Applying LLMs to Recommendation Experiences

Challenges and lessons from deploying LLM experiences: evals, scalability, guardrails.

#3000:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
31
Prompting Fundamentals and How to Apply them Effectively

Structured input/output, prefilling, n-shots prompting, chain-of-thought, reducing hallucinations, etc.

#3100:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
32
What We've Learned From A Year of Building with LLMs

From the tactical nuts & bolts to the operational day-to-day to the long-term business strategy.

#3200:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
33
Building an AI Coach to Help Tame My Monkey Mind

Building an AI coach with speech-to-text, text-to-speech, an LLM, and a virtual number.

#3300:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
34
Task-Specific LLM Evals that Do & Don't Work

Evals for classification, summarization, translation, copyright regurgitation, and toxicity.

#3400:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
35
Don't Mock Machine Learning Models In Unit Tests

How unit testing machine learning code differs from typical software practices

#3500:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
36
How to Generate and Use Synthetic Data for Finetuning

Overcoming the bottleneck of human annotations in instruction-tuning, preference-tuning, and pretraining.

#3600:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
37
Language Modeling Reading List (to Start Your Paper Club)

Some fundamental papers and a one-sentence summary for each; start your own paper club!

#3700:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
38
2023 Year in Review

An expanded charter, lots of writing and speaking, and finally learning to snowboard.

#3800:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
39
Push Notifications: What to Push, What Not to Push, and How Often

Sending helpful & engaging pushes, filtering annoying pushes, and finding the frequency sweet spot.

#3900:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
40
Out-of-Domain Finetuning to Bootstrap Hallucination Detection

How to use open-source, permissive-use data and collect less labeled samples for our tasks.

#4000:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
41
Reflections on AI Engineer Summit 2023

The biggest deployment challenges, backward compatibility, multi-modality, and SF work ethic.

#4100:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
42
AI Engineer 2023 Keynote - Building Blocks for LLM Systems

Evals, retrieval-augmented generation, guardrails, and collecting feedback; all that good stuff.

#4200:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
43
Evaluation & Hallucination Detection for Abstractive Summaries

Reference, context, and preference-based metrics, self-consistency, and catching hallucinations.

#4300:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
44
How to Match LLM Patterns to Problems

Distinguishing problems with external vs. internal LLMs, and data vs non-data patterns

#4400:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
45
Patterns for Building LLM-based Systems & Products

Evals, RAG, fine-tuning, caching, guardrails, defensive UX, and collecting user feedback.

#4500:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
46
Obsidian-Copilot: An Assistant for Writing & Reflecting

Writing drafts via retrieval-augmented generation. Also reflecting on the week's journal entries.

#4600:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
47
Some Intuition on Attention and the Transformer

What's the big deal, intuition on query-key-value vectors, multiple heads, multiple layers, and more.

#4700:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
48
Open-LLMs - A list of LLMs for Commercial Use

It started with a question that had no clear answer, and led to eight PRs from the community.

#4800:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
49
Interacting with LLMs with Minimal Chat

Should chat be the main UX for LLMs? I don't think so and believe we can do better.

#4900:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入
50
More Design Patterns For Machine Learning Systems

9 patterns including HITL, hard mining, reframing, cascade, data flywheel, business rules layer, and more.

#5000:00 达到00:00首次采集时已在榜—结束时仍在榜7小时37分累计在榜1无重入