惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

F
Fortinet All Blogs
Microsoft Security Blog
Microsoft Security Blog
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
Vercel News
Vercel News
Application and Cybersecurity Blog
Application and Cybersecurity Blog
C
Check Point Blog
cs.AI updates on arXiv.org
cs.AI updates on arXiv.org
W
WeLiveSecurity
The Hacker News
The Hacker News
L
LINUX DO - 热门话题
T
Tenable Blog
Hugging Face - Blog
Hugging Face - Blog
Google Online Security Blog
Google Online Security Blog
博客园 - Franky
P
Proofpoint News Feed
H
Hacker News: Front Page
P
Privacy & Cybersecurity Law Blog
月光博客
月光博客
P
Proofpoint News Feed
Exploit-DB.com RSS Feed
Exploit-DB.com RSS Feed
The GitHub Blog
The GitHub Blog
云风的 BLOG
云风的 BLOG
博客园_首页
www.infosecurity-magazine.com
www.infosecurity-magazine.com
C
CERT Recently Published Vulnerability Notes
Forbes - Security
Forbes - Security
I
InfoQ
Stack Overflow Blog
Stack Overflow Blog
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
Attack and Defense Labs
Attack and Defense Labs
N
News and Events Feed by Topic
博客园 - 叶小钗
T
Threat Research - Cisco Blogs
aimingoo的专栏
aimingoo的专栏
D
Darknet – Hacking Tools, Hacker News & Cyber Security
小众软件
小众软件
大猫的无限游戏
大猫的无限游戏
MongoDB | Blog
MongoDB | Blog
CTFtime.org: upcoming CTF events
CTFtime.org: upcoming CTF events
Hacker News - Newest:
Hacker News - Newest: "LLM"
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
博客园 - 司徒正美
O
OpenAI News
G
Google Developers Blog
Martin Fowler
Martin Fowler
罗磊的独立博客
S
SegmentFault 最新的问题
T
Tor Project blog
量子位

Hacker News - Newest: "AI"

AI can't read an investor deck AI as an attorney? Student uses ChatGPT, Gemini to sue UW over alleged racial discrimination Hacking MCP Servers in AI Systems – The Rug Pull: Tool Changes After Approval GitHub - MeepCastana/KubeezCut: Free Web based video editor GitHub - GenAI-Gurus/awesome-eu-ai-act: Curated tools, official sources, OSS, templates, and guides for EU AI Act compliance. Can AI judge journalism? A Thiel-backed startup says yes, even if it risks chilling whistleblowers Coming soon: 10 Things That Matter in AI Right Now DARPA built an AI to fact-check enemy weapons claims What explains heterogeneity in AI adoption? When AI Meets Muscle: Context-Aware Electrical Stimulation Promises a New Way to Guide Human Movements - Department of Computer Science AI Changed How We Build. It Did Not Change What Matters. Linux rules on using AI-generated code - Copilot is OK, but humans must take 'full responsibility for the… Meta spins up AI version of Mark Zuckerberg to engage with employees Code Mode: Let Your AI Write Programs, Not Just Call Tools | TanStack Blog GitHub - Delavalom/graft: Go framework for building AI agents. Type-safe tools, multi-provider (OpenAI, Anthropic, Gemini, Bedrock), zero vendor SDKs. India's TCS tops estimates, says new AI models did not dent services demand Gen Z's fading AI hype Strong feeling: we are in a folded AI reality GitHub - machinarii/total-recall-catalog: A reference catalog of latest knowledge retrieval, memory & RAG systems GitHub - mensfeld/code-on-incus: Give each AI agent its own isolated machine with root, Docker, and systemd. Active defense detects and stops threats automatically.. Quantization, LoRA, and the 8% Problem: Benchmarking Local LLMs for Production AI Iran war: We spoke to the man making Lego-style AI videos that experts say are powerful propaganda Powell, Bessent discussed Anthropic's Mythos AI cyber threat with major U.S. banks GitHub - immartian/bellamem: Persistent belief-graph memory for AI agents. Retrieves decisive context by importance — not recency, not RAG, not /compact. recursive-mode: The Repo-Native Operating System for AI Engineering After the attack on Sam Altman's home, will AI CEO's go on the offensive? The biggest advance in AI since the LLM Opus 4.6 vs GPT 5.4 One Prompt Unity World Generation Test “AI polls” are fake polls Client Challenge Can AI be a 'child of God'? Inside Anthropic's meeting with Christian leaders How to Switch AI Chatbots and Why You Might Want To GitHub - MattMessinger1/agentic_refund_guardrail: Safe refund policy layer for AI agents — Python + TypeScript. Same behavior, shared tests. Adam/papers/emergent_values_whitepaper.md at master · strangeadvancedmarketing/Adam Ask HN: How do you stop playing 20 questions with your AI coding tools How far can automation and AI support psychotherapy? - @theU GitHub - stagas/rtdiff: realtime git diff gui and AI-assisted commits A Mac Studio for Local AI — 6 Months Later A History of the Early Years of AI at the University of Edinburgh Why AI Coding Tools Still Feel Stuck on Localhost MSN AI Datacenters Are Becoming Strategic Targets twitter.com Penn Researchers Use AI to Surface Unreported GLP-1 Side Effects in Reddit Posts Show HN: MoodSense AI (ML and FastAPI and Gradio, Deployed on Hugging Face) Moodsense Ai - a Hugging Face Space by aman179102 AI models are terrible at betting on soccer—especially xAI Grok GitHub - xialeistudio/echoic GitHub - HimashaHerath/github-dev-wrapped: AI-powered weekly GitHub activity reports deployed to GitHub Pages GitHub - alejandrobalderas/claude-code-from-source: Architecture, patterns & internals of Anthropic's AI coding agent — reverse-engineered from source maps AI and Tech brief: Ireland ascendant GitHub - Titovilal/context0: Context0 - Never Surrender Training for a Marathon with an AI Coach: What Worked and What Didn't Cyber Pulse: Agentic Intel - Apps on Google Play I Built an AI PR Reviewer That Catches Bugs by Not Looking for Bugs Gen Z workers are so fearful AI will take their job they’re intentionally sabotaging their company’s AI rollout | Fortune How AI Is Reimagining the Game of Golf–For Both Players and Courses GitHub - nattergabriel/reseed: A CLI tool for managing and distributing agent skills across projects Is SVG the final frontier? My AI workflow evolved from prompts to a near-autonomous workflow MLSharp Help - 3DGS Viewer & Generator I put my cognitive field based AI's runtime on GitHub Is Numble the first AI-proof game? A3: Kubernetes for autonomous AI agent fleets | Emergent Principles Deepali Vyas ("The Elite Recruiter") GitHub - msmarkgu/RelayFreeLLM: A restful API designed to route user prompts to various AI model providers. Unionized ProPublica staff are on strike over AI, layoffs, and wages Unleashing the Advantage of Quantum AI We're heading for an AI-fueled 'dementia crisis,' brain scientist warns The AI-Assisted Breach of Mexico's Government Infrastructure [pdf] GitHub - stef41/lmscan: 🔍 Detect AI-generated text and fingerprint which LLM wrote it. Open-source GPTZero alternative. Zero dependencies, works offline. MSN GitHub - visionscaper/collabmem: Enabling long-term collaboration with Agentic AI - building up episodic and world model memory over time with in-context awareness We gave an AI a 3 year retail lease in SF and asked it to make a profit | Andon Labs AI Code is Hollowing Out Open Source, and Maintainers are Looking the Other Way What leaked "SteamGPT" files could mean for the PC gaming platform's use of AI AI is the boss at this retail store. What could go wrong? GitHub - Wuzu11517/agentic-proxy: Local proxy meant to help reduce With Drones, Geophysics and ArtificiaI Intelligence, Researchers Prepare to Do Battle Against Land Mines A Single Operator, Two AI Platforms, Nine Government Agencies: The Full Technical Report 在 Steam 上购买 FriedrichAI: Offline AI 立省 10% GitHub - inevolin/resume-cli: Hit Claude usage limits? Resume any AI coding session elsewhere. Switch tools at zero friction. GitHub - atripati/ark: AI Runtime Kernel — a context operating system for AI agents. Eliminates tool bloat, loads only what’s needed, and gives LLMs their reasoning space back. How to Build a Secure AI PR Reviewer with Claude, GitHub Actions, and JavaScript This Startup Wants You to Pay Up to Talk With AI Versions of Human Experts Intel Arc Pro B70 Brings 32GB VRAM to Local AI for $949 WordPress 7.0: The Good, the AI, and the Still Missing AI on the couch: Anthropic gives Claude 20 hours of psychiatry IatroBench: Pre-Registered Evidence of Iatrogenic Harm from AI Safety Measures AI Agents Know About Supabase. They Don't Always Use It Right. The history and future of AI at Google, with Sundar Pichai Inside an AI‑enabled device code phishing campaign How Meta Used AI to Map Tribal Knowledge in Large-Scale Data Pipelines AI for Systems: Using LLMs to Optimize Database Query Execution Forecasting the Economic Effects of AI Introducing Tinker: Play with AI, bring your ideas to life AI sheds light on an ancient gaming mystery People really hate AI but not as much as Iran—or Democrats | Fortune What is an AI Product Engineer? Phoebe Gates wants her $185 million AI startup to succeed with 'no ties to my privilege or my last name': 'I have a chip on my shoulder' | Fortune
Being AI-native matters more than experience - PostHog
duck · 2026-05-22 · via Hacker News - Newest: "AI"

In Q2 2026, our Ops & Finance team caved and said what we'd been thinking for a while: when it comes to hiring, AI-native beats experience.

Not "experienced person who uses AI." Not "senior hire with AI on the JDs as a nice-to-have." AI-native, full stop. Years of experience are now the tiebreaker between two AI-native candidates, not the gate that decides who gets to the table.

At our Q2 planning, we made the decision that every team member needed to be using agents and LLMs to improve their workflows. This is where the AI-native person cooks.

AI-native sounds lovely, but it's also a buzzword. This post is meant to show you what it means in practice and how we are implementing it into Ops & Finance at PostHog.

Standard hiring advice for ops roles is still in the pandemic era. Find people who know what they are doing. Except now, you have to put “comfortable working with AI” in your job descriptions. Why else did we all sign off those huge AI budgets, right? If you're good at your job and you use AI to be more productive, great – you're probably 20% faster than the colleague who doesn't.

Being AI-native isn't just a layer on top of a job. It's a totally different version of it. The AI-native accountant doesn't do accounting + AI. They produce the same work but 5x faster and with way better controls in place.

We aren't just letting anybody through the door. The person still needs to actually be an [insert job title], they can't just be an AI junkie who fancies being an accounting manager.

From an Ops perspective, we have three examples:

1. Ops manager

A traditional excellent ops manager is a hero. They are everywhere. They are the glue. This person is successful because they care, are reliable, and can go really wide on tasks. They think of everything that needs to go on an onboarding checklist, tag the right people, and chase them up continually until it's done. However, things fall through the cracks and they get blocked whilst waiting for others.

The AI-native ops manager spins up a Slack bot that pushes tasks to owners, nudges them when something's due, and escalates when it's overdue. They build an offer-letter generator for the talent team that hooks into employment contract generation across the 30+ countries PostHog hires in — fully automated, end to end.

Shoutout Tara Alcantarilla-HowardTara Alcantarilla-Howard and Carol DonnellyCarol Donnelly.

2. Talent partner

A traditional talent partner is always on. They're on the frontline internally and externally, managing lots of people, having difficult conversations, and smiling through all of it. They work late doing calls to find the right person, chase feedback from hiring managers, and reply to rejected candidates at 11pm. They often burn out quickly.

The AI-native talent partner creates a feedback summary generator that takes all the feedback on why we won’t pass somebody, runs it through an agent that knows our tone of voice and delivers solid, friendly and accurate feedback to candidates. They use the Ashby MCP server to pull all the qualitative feedback on our strongest performing hires and run analysis in Claude to show us the trends of people that do great at PostHog. They also run the same analysis on hires that have not worked out and how that matched the feedback from our hiring process.

Did you know that people who talk fast usually really succeed at PostHog? Talk fast, think fast. We found that out from an AI-native talent partner.

Shoutout Rune PovlsenRune Povlsen.

3. Finance manager - accounting

On day 30, a traditional finance manager has created a nice close checklist, spent 3 days reviewing last month’s financials, and written a management report by hand that needs proof-reading before going out to the company. They spend more time creating financials than helping the business make the content of them better.

The AI-native finance manager, by day 30, has a recurring close-checklist agent that can read the GL, flag variances over a set amount, and post them direct to Slack. They’ve also got your ERP hooked up to an MCP and has a dope skill that creates the most fire finance update you’ve seen. They’ve also knocked up an accounts receivable agent that is sending personalized emails following up with overdue invoices and tagging the right person when it needs a human. Cash in the bank quicker.

Shoutout Janani KJanani K and Ahmed AmaarAhmed Amaar.

These roles don't create some kind of AI-eutopia, there are still risks which we need to take seriously.

  • When it comes to accounting, you still need qualified accountant signoff, and we still need to pass an audit, so we still rely on years of experience and certain credentials.

  • We also only consider people that also show good judgement when it comes to things like relationships and taste. These things are particularly important at PostHog.

  • There is a risk you fill the team with people who can’t function without the tools at their disposal, this over-reliance on AI is also a failure mode.

  • AI-native and AI-curious are very different. Plenty of people have a Claude subscription and can prompt a chat but don’t understand workflows and artifacts. This differential is important, we’ll talk about how to screen for this below

Everybody says they are AI-native so how do you actually find the cracked people amongst the noise. The PostHog hiring process is more or less split in two. We have three interviews that are all about if you can do all of the above in theory.

We ask questions like:

  • Walk me a through a workflow you built or replicated: what does it do? what was the aim for it? how did it improve things (faster, more accurate, less human interaction)?

  • What’s the best AI workflow you’ve stolen from somebody else? Nobody actually knows what we are doing but we all steal the cool stuff from each other.

  • What’s the biggest failed experiment you’ve had with AI? We assess how much people are pushing themselves. We’ve all failed somewhere, the first board pack my agents created would have had PostHog shutdown.

After these first 3 stages, we do our SuperDay. This shows that the person can do the role in practice. It involves them building something for us during the day.

We recently hired for a payroll manager and we asked candidates to build a workflow to deliver a flux analysis on payroll runs. The great submissions built a proper workflow that allowed you to input the two Excel files of payroll data and get an output flagging the errors. This could be a Python script or a web app, we just wanted something that produced a repeatable process that a non-payroll expert can use. The weak submissions were screen recordings of someone asking ChatGPT basic questions.

It’s important at this stage to get the details. You want to see the prompts, the skills files, info on the agents. The details here matter. You also want to make sure they can explain this to you via some kind of recording or live in a call. Fluency of tools is also important as we’re not testing on a specific stack. It is a strong signal if the person is using multiple different tools, rather than gating themselves on one, it shows they’ve pushed themselves to learn more.

Conventional wisdom says ops headcount is about 8-10% of a well run company. PostHog currently operates with it at 6% and we don't think we need to grow the team anytime soon. We feel we can do much more with less now.

The practical result: PostHog's hiring plan for the next 2-3 years is smaller than it would have been a year ago. Not because we're growing slower. Because each AI-native hire is cracked enough that we need fewer of them.

We don't know yet what people will do with the time saved by AI. The optimistic version is the 3 hours a day you save get reinvested into building, learning the fundamentals, and finally writing your own Python scripts. The pessimistic version is Parkinson's law eats it and we all just have more meetings.

We're betting on the optimistic version. The AI-native person has just had an explosion of resourcefulness, and we want a whole company of them.

If that sounds like fun, believe it or not, we're actually still hiring.