Urgent.News

What's breaking now, across thousands of outlets.

Saturday, 15 August 2026 Reading Room

Tech

Intel's Jay Graphics Shader Compiler Now Passing Vulkan CTS With Xe2/Xe3 Hardware

Last year Alyssa Rosenzweig joined Intel to work on their Linux graphics drivers after previously serving as a contractor for Valve on their graphics stack and also previously leading the Mesa driver work for the Asahi Linux Apple Silicon graphics support. Since joining Intel, she has been leading the development of Jay as a new SSA-based shader compiler for Intel graphics hardware to be used by…

phoronix.com ·

We haven't written up this one. Phoronix has the full story — the link below goes straight to it.

Read the original at phoronix.com →

More in Tech

Where the time actually goes in an AI coding-agent job

I run an agent-orchestration platform: it takes a ticket, spins up a git worktree, lets a Claude Code agent build the feature, and runs a deterministic verify gate before merging.

  • Infrastructure costs dominate AI coding-agent job time, not model inference
  • 20-30% of total job time attributed to infrastructure overhead
  • Optimizations can reduce overhead to under 60 seconds, leaving model reasoning as key phase

·dev.to

The Twenty-Hour Fuse

Two nights of debugging, one incident left to write about. The night of August 14th, a Claude Code session hung for an hour inside the VS Code extension, and I traced it to a per-token streaming stall…

  • Claude Code session froze for an hour on August 14th
  • Server instability caused by high load average of 38.7
  • Llama.cpp process consumed excessive memory and triggered IO storm

·dev.to

Coalescing the Stream

Claude Code, on my box, doesn't talk to a model provider directly. It goes through claude-code-router (ccr) — a self-hosted proxy that lets the same CLI point at the official Anthropic API or at…

  • Claude Code uses claude-code-router to switch between Anthropic API and third-party providers.
  • Middleware merges streaming events into chunks without altering stream meaning.
  • Middleware maintains content-length header and avoids compression on merged bodies.

·dev.to

Measuring the real concurrency ceiling of an LLM agent runner

I wanted to raise the concurrency limits on my local AI agent runner. The UI now supports multiple terminal panes running in flight, and my gut told me the runner process itself was becoming the…

  • Initial belief that runner process bottlenecked performance
  • Benchmark revealed Ollama as the actual performance bottleneck
  • Scheduler hardcoded one job per repo, limiting concurrency

·dev.to

I built 49 free online tools in 3 months — here's what I learned

I have a confession. Every time I needed to calculate my EMI, check my BMI, or figure out how many college classes I could skip — I'd Google it, land on some site plastered with ads, get asked to sign…

  • Developer created 49 free online tools in 3 months
  • Tools cover finance, health, freelance, student, and utility categories
  • Rules: no signups, no ads, mobile compatibility

·dev.to

More from Saturday 15 August →