The circular economy of talent in AI might be the best I've ever seen. > Anthropic has 55 engineers who came from OpenAI > OpenAI has 14 from Anthropic > Stripe is Anthropic's third-biggest feeder with 98 engineers > 71 of DeepMind alumni now build at OpenAI, 23 at xAI > xAI is a taker. It took 19 from OpenAI, and 23 from DeepMind

We don't make candidates do any LeetCode in our hiring. A candidate just shared their experience with Chakra, our agentic interviewer. It spent 60 minutes with them on: > Fixing a broken production system > Measuring real user impact > Weighing system trade-offs Anyone can memorize syntax. Engineers should be evaluated on real-world judgment and how they work with AI.

More than 500 companies are already using AI in their hiring process with HackerRank. And in 2026, more than 260,000 candidates have taken an AI-assisted assessment on our platform. Is your team hiring for the agentic era?

Cursor’s final round is just you doing the job for two days. You get a desk, a Mac, the dev environment already set up, and a few actual projects to choose from. You pick one, build it, and demo it the next day. You also eat 4-6 meals with the team. This isn't scalable, but work trials most certainly give the best signal on technical proficiency and cultural fit.

This guy vibecoded all his SaaS. It went well! 👍

LeetCode is dying. A candidate recently interviewed at 6 companies. 4 of them had completely ditched LeetCode. Instead of inverting binary trees, they were asked to: — Build features in a real codebase — Use AI to navigate unfamiliar code — Collaborate like actual coworkers The 45-minute DSA grill is slowly coming to an end. This is the shift in modern software engineering hiring. It is time to mirror the interview process with how you build.

Founders in 2022: make products with good foundation. Founders in 2026: make videos with good foundation.

An engineer landed an L4 SWE offer at Anthropic without solving a single LeetCode problem. This is because top AI companies no longer care about algorithm memorization. While I expect some level of algorithmic knowledge to be relevant, the weightage will increasingly shift to real-world tasks that evaluate your judgment, nd taste vs rote memorization.

Varun asked for an interview if they hit Top 20 in our Orchestrate hackathon on Aug 1st. And that's a deal. But I'll go one further. Anyone who reaches Top 20 gets an interview. If there's a fit, we make an offer.

Three hiring filters that matter way less than people think: 1 - GPA 2 - Your University 3 - Years of experience Three hiring filters that matter far more than people think: 1 - Subscribing to the company's mission 2 - Learning curve 3 - High agency

Cursor’s founder on why AI makes the next engineer more valuable, not less. The New York Times spends ~$150M-$200M a year on software R&D, even though most people do not think of it as a software company. Professional software engineering is still far from solved because building a new codebase is not the same as safely migrating an existing system like Rippling’s 30M-line codebase, with years of logic, dependencies, and customer workflows behind it. Cursor’s 20-person support team handling millions of daily users clearly shows that AI leverage for small teams is real. But the scope of software work is still vast, so the ROI of adding the next strong engineer is higher, not lower.

If you're struggling to land your next job, do this: > stop grinding LeetCode all day > learn to work with agents > build cool stuff, post it publicly, and tag the CEO > don’t let AI mass apply for you > curate 10 strong applications instead of sending 100 lazy ones > cold email with what you can do, what you’ve done, and what problem you’ll solve for them Steve cold emailed me like this. Now he works at HackerRank. It works. I promise.

If you unfortunately get laid off, don’t just write the “I was impacted” LinkedIn post. But also create value out of what you learned. Vasilios Syrakis was laid off from Atlassian after 8 years. Instead of only posting that he was open to work, he made a 40-minute video explaining the systems he had worked on. That video went viral. He got heavy inbound interest. His outreach started getting “100% yes.” And now when a recruiter searches his name, that video becomes the first impression. If you get laid off, teach the system you built, it might not go as viral, but it'd turn your work into proof. And I'd hire a person like that.

Got called a lightweight Cursor by a candidate on Reddit. Should I be ready for my $60B acquisition @elonmusk? Anyway, glad to see technical interviewing changing!

Cursor's way of making its first 20 hires was through community. They started with 20 users they hand-onboarded one by one. Most of them actually left, they didn't like it, and a few even ghosted. But a couple of people stayed, and that became a community in a Discord server. Cursor hired many of its early team straight out of that server. Early users as hires are very powerful.

Both are very talented people. God Bless!

Nvidia is one of the most valuable companies on earth. But it still does not manufacture its own chips. Its future depends on a company in Taiwan that makes the chips, and a company in Veldhoven that makes the machines needed to make them. This is the hidden supply chain behind AI. 0:00 - The Nvidia Contradiction 0:35 - Chapter 1: The Design 1:42 - Chapter 2: The Manufacturing 2:59 - Chapter 3: The Machine 3:58 - Chapter 4: The Data Center 5:02 - Chapter 5: What This Means for the Future of AI

Who builds the future of AI?