Coding agents

GitHub repository card for pingdotgg/ts-rust, described as an experimental Rust port of the TypeScript 7 compiler

An LLM-driven Rust port of TypeScript reports 181,711 passing tests

Oct 8, 2026

Linear chart shows 3.7 times as many tests and 46 percent less machine time per test; Linear

AI coding moved Linear's bottleneck from generation to verification

Sep 22, 2026

A modular tool arm stands behind green, amber, and red translucent permission gates

GitHub adds enterprise-managed permissions for Copilot agent operations

Sep 11, 2026

An enterprise AI coding policy panel routes model access, billing seats, and review controls

GitHub changes Copilot policy and billing for enterprise AI work

GitHub announced Copilot policy and billing changes covering paid seat assignment, usage overages, cloud-agent convergence, data retention, and code-review defaults.

By The AI Feed Desk

Aug 30, 2026

A model access gate blocks a stream of coding-workflow tiles beside a cutoff calendar

OpenAI says Cursor will lose model access after SpaceX acquisition

OpenAI told SpaceX it intends to wind down Cursor's OpenAI model contract on November 12, 2026, turning the Cursor acquisition into a developer access-risk story.

By The AI Feed Desk

Aug 30, 2026

A bright workbench shows a fast model core connected to code files, browser panels, and agent task steps

Google makes Gemini 3.7 Flash GA for coding agents

Google's Gemini API release notes list Gemini 3.7 Flash as generally available for software engineering, web development, and agentic workflows.

By The AI Feed Desk

Aug 21, 2026

A developer workbench shows completed agent requests with model labels, credit meters, cache indicators, and session branches

GitHub Copilot Auto now shows model and AI-credit details

GitHub Copilot's weekly release adds model, AI-credit, and cache visibility for Auto requests plus new CLI session controls.

By The AI Feed Desk

Aug 9, 2026

A coding agent runs in isolated sandboxes while a training loop records tool-call tokens and hidden-test rewards

Hugging Face shows loop-owning training for coding agents

Hugging Face demonstrated a TRL and OpenEnv workflow that trains on the tokens produced by a real OpenCode agent loop in remote sandboxes.

By The AI Feed Desk

Aug 7, 2026

A coding model switchboard routes an open-weight model through policy gates and an incident pause marker

GitHub pauses Kimi K3 Copilot rollout during Actions incident work

GitHub added Kimi K3 to Copilot across major surfaces, then temporarily paused the rollout while mitigating a GitHub Actions incident.

By The AI Feed Desk

Aug 7, 2026

A security workbench inspects cracked and complete code patch tiles under a magnifying lens

1Password finds AI vulnerability patches often leave defects

1Password's Off-by-1 Labs tested 6,080 AI-generated patches across six recent CVEs and found that complete fixes arrived only 26.0% of the time.

By The AI Feed Desk

Aug 6, 2026

A comment trigger, reasoning dial, and pull request checkpoint connect through a cloud agent workspace

GitHub gives Copilot cloud agent comment triggers and reasoning controls

GitHub added comment-triggered Copilot automations and per-task reasoning controls for Copilot cloud agent.

By The AI Feed Desk

Aug 4, 2026

Metrics, logs, and traces converge toward an unresolved root-cause node on an abstract observability dashboard

ORCA-bench shows coding agents are not ready for on-call root cause analysis

A new benchmark tests AI agents on on-call root cause analysis with metrics, logs, traces, code, and 1,079 incident tasks.

By The AI Feed Desk

Aug 3, 2026

A luminous Q-shaped data path runs between cloud infrastructure and modular open model weights

Alibaba releases Qwen3.8-Max with QwenCloud access and open weights next week

Qwen3.8-Max puts Alibaba back into the frontier-model fight with a large coding-focused model available through QwenCloud now and open weights promised next week.

By The AI Feed Desk

Aug 3, 2026

A compact security model routes vulnerability tasks through multiple agent lanes toward a protected codebase

Microsoft routes vulnerability agents through MAI-Cyber-1-Flash

Microsoft introduced MAI-Cyber-1-Flash inside MDASH, saying the compact security model handles most vulnerability tasks while larger models handle harder cases.

By The AI Feed Desk

Aug 2, 2026