Developer Tools
Claude Code Adds Agent View for Simplified Session Management
Claude Code's new Agent View streamlines managing multiple sessions, offering better control and efficiency for developers.
GitHub Details Outage Causes and Scaling Plans Amid Rapid Growth
GitHub outlines steps to enhance reliability following recent outages affecting pull requests and search systems. Scaling plans aim to meet soaring demand.
GitHub Pauses Copilot Signups as AI Agents Overwhelm Infrastructure
GitHub halts new Copilot Pro and Pro+ subscriptions, tightens usage limits, and removes Opus models from cheaper tiers as agentic AI workloads strain capacity.
OpenAI Agents SDK Gets Sandbox Execution and Model-Native Harness
OpenAI releases major Agents SDK update with native sandbox execution and enhanced harness for building secure, long-running AI agents across files and tools.
GitHub Overhauls Status Page With New Degraded Performance Tier
GitHub introduces three-tier incident classification, per-service uptime metrics, and dedicated Copilot AI model tracking in major status page transparency update.
OpenAI Codex Update Adds Computer Control for 3M Weekly Developers
OpenAI releases major Codex update with background computer use, 90+ plugins, and memory features for over 3 million developers using the coding assistant weekly.
Anthropic Unveils Claude Opus 4.7 with New Adaptive Thinking for Developers
Anthropic's Claude Opus 4.7 introduces adaptive thinking and recalibrated effort levels, promising better code review and reduced token usage for enterprise workflows.
Anthropic Unveils Claude Code Session Tools for 1M Token Context
Anthropic releases /usage command and context management guide for Claude Code, helping developers navigate 1 million token sessions more effectively.
NVIDIA NVbandwidth Tool Gets Multi-Node Support for AI Infrastructure Testing
NVIDIA's NVbandwidth benchmarking tool now supports multi-node GPU clusters, enabling developers to measure bandwidth across NVLink connections at 397+ GB/s.
LangChain Warns AI Agent Memory Lock-In Could Create Vendor Monopolies
LangChain argues closed AI agent harnesses create dangerous vendor lock-in through proprietary memory systems, pushing developers toward open-source alternatives.