We're building assessments that feel like the actual job, with the tools you'd use on it. x.com/rvivek/status/19933558…

Last month thousands of developers from different countries showed up to build AI agents. Now we're back. HackerRank Orchestrate June edition is here to challenge the next generation of AI-native builders. June 19 - 20 | Registrations close 18th June, 7:30 AM PT Register now: hackerrank.com/hackerrank-or…

Developers aren't just writing code anymore. They're orchestrating AI agents, reviewing AI-generated work, and deciding what's safe to ship. The job has changed. The interview process has to keep up. That's what our CEO @rvivek is talking about at @FD1conference, Manhattan on June 16. How leading companies are rethinking technical hiring for the agentic era, and what it actually takes to make the shift. Join us at Columbia's Lerner Hall: fromdayone.com/conferences/2…

Agents are getting better at doing the work. They’re still bad at knowing whether the work is done. That’s one of the hardest problems in long-running agents. They can confidently stop after building half a feature. The fix isn’t just "give it more context." Two patterns matter more, and most teams skip them: 1. Don’t let the builder agent grade its own homework. Use a separate evaluator agent to check the output against clear acceptance criteria. 2. Make failure predictable. Break work into bounded loops with four constraints: Fresh context for each loop, max iterations, stop conditions, and a clear exit criteria. Trustworthy agents need a verifier and a clock. Most teams ship with neither.

Claude Fable 5 is too good. Just made a Claude clone with it. Watch out, Anthropic 👀

If AI Models were people - Ep1 featuring Claude Code, Codex, Cursor, and Deepseek🤫

Who builds the future of AI?

Where's the lie?

Anyone can prompt. Few can build systems that actually deliver. HackerRank Orchestrate is where developers prove they can do the second one. 24 hours. Real-world problem. Ship an AI agent that works. June 19 - 20 | Virtual | Registrations close June 18, 8 PM IST Register now: hackerrank.com/hackerrank-or…

Hire for the agentic era.

More MCP tools will just make your agent worse. When your MCP has hundreds of tools to choose from it becomes a liability. The model has to read every tool definition. Choose the right one, pass the right inputs, and not blow up the context window before the work even starts. The better pattern isn't giving the agent more choices. It's giving it a cleaner way to find the right one. Instead of giving the agent hundreds of preloaded tools, give it a smaller interface: search what exists, write the code, then execute the right call.

Behind every great team is someone who made the harder, better call. Someone who looked past the resume and hired for what people could actually do. The HackerRank Innovator Awards is the night that work gets recognized. 📍 The Ritz-Carlton, Bengaluru | June 12, 2026 Register now: hackerrank.com/events/innova…

Developer wakes up from a coma after 2 years