Your source for technology insights, tutorials, and guides.
Explore seven key facts about Bun after its Anthropic acquisition, covering speed, memory issues, maturity, and future viability for developers.
Airbyte Agents precomputes business data into a unified index, reducing AI agent API calls from six to one, solving latency and token cost issues for production deployments.
Subquadratic's new 12M-token model uses linear-attention SSA architecture, outperforming GPT-5.5 on MRCR v2 and SWE-bench, with plans for a 50M-token version.
Amazon expands AI coding tools to Claude Code and Codex after developer pushback, ending Kiro-only mandate and transforming developer workflows.
A step-by-step guide to finding and using CSS color palettes without Tailwind, covering pre-made palettes, generators, modern color functions, accessibility testing, and integration via custom properties.
V8's optimization of mutable heap numbers eliminated allocations in Math.random, delivering a 2.5x improvement in the async-fs benchmark.
V8 replaces Sea of Nodes with CFG-based Turboshaft in Turbofan, solving maintainability issues and simplifying compilation, with only remnants of SoN remaining.
Learn how Chrome 136's Explicit Compile Hints feature reduces JavaScript startup time by allowing developers to mark files for eager compilation. Includes experiments, implementation, and cautions.
Explains V8's new speculative inlining and deoptimization for WebAssembly, boosting WasmGC performance significantly, with over 50% speedup on Dart microbenchmarks and 1-8% on real apps.
V8's recent optimization more than doubles JSON.stringify speed via a side-effect-free fast path, iterative traversal, and templatized string handling. Learn the technical details in this Q&A.
TechCrunch Disrupt 2026 offers 50% off a second pass for 4 more days. Prices increase on May 8. Act now to bring a colleague at half price.
ASML CEO Christophe Fouquet asserts no competitor threatens its lithography monopoly, citing technological lead and R&D investment.
Manufacturing shifts from physical testing to simulation-first using OpenUSD and SimReady. ABB achieves 99% sim-to-real accuracy; JLR cuts aerodynamic sim time from 4 hours to 1 minute.
NVIDIA's Nemotron 3 Nano Omni unifies vision, audio, and language in an open model, delivering 9x higher throughput and top accuracy for multimodal AI agents.
GeForce NOW adds 16 games in May, including Forza Horizon 6 and 007 First Light, plus RTX 5080-class performance expansion for Ultimate members.
OpenClaw, a self-hosted persistent AI assistant, became GitHub's top project in 2025. NVIDIA collaborates to enhance security and introduces NemoClaw for safe enterprise deployment.
NVIDIA and ServiceNow partner to launch autonomous AI agents for enterprises, featuring Project Arc, OpenShell secure runtime, and domain-specific models.
A step-by-step tutorial on using PaletteInspiration.com's 3,000+ master painter palettes and Color Harmony Explorer to find historically informed color combinations for design projects, with export and application tips.
Limited-time 4-day offer: Buy one TechCrunch Disrupt 2026 pass, get 50% off a second. Ends May 8 at 11:59pm PT. Act now to save on bringing a colleague.
5 key takeaways from ASML CEO Christophe Fouquet's interview: competition, geopolitics, talent, Moore's Law, and leadership philosophy.