WideSWE: Can Coding Agents Synchronize Modifications Across Multiple Repositories?
WideSWE evaluates coding agents on 120 real multi-repo tasks, showing low full success and failures in discovery, execution, and completeness.
AI coding agents
The Rise of AI Agents: From AutoGPT to Claude Code
AI coding agents have moved beyond AutoGPT-style autonomous loops toward narrower, workflow-aware tools like Claude Code, Codex, Cowork, and OpenClaw. The key insight is that agent value now depends o…
Building Python Agents: From Intuition to Evidence
Agentic engineering helps Python devs use AI coding agents safely through bounded tasks, two loops, evidence checks, RECAP review, and acceptance.
NVIDIA Unveils SoL-Pi: Self-Driving Research Cycles That Reduce Coding Agent Token Usage by as Much as 49%
NVIDIA, NTU, and MIT researchers released SoL-Pi, an MIT-licensed extension for the open-source Pi coding agent that cuts recorded token traffic by 44.7–49% on EdgeBench and lowers API costs by roughl…
Say Hello to GPT-5.3-Codex
GPT-5.3-Codex is a Codex-native AI coding agent built from the ground up for long, real-world engineering tasks. Unlike chatbots adapted for code, it combines frontier-level coding performance with ge…
Creating a secure and efficient sandbox environment to bring Codex support to Windows
OpenAI brought Codex to Windows by building a native sandbox using restricted tokens and Job Objects. Restricted tokens limit child processes to the workspace folder via ACLs, while Job Objects contai…
Warp makes a major push into open source development with GPT-5.5
Warp has evolved from a modern terminal into an intelligent development environment powered by GPT-5.5 and other OpenAI models. It enables developers to launch, manage, and coordinate multiple AI codi…
How Codex helped Wasmer create a Node.js runtime for edge environments
Wasmer leveraged Codex with GPT-5.5 to build a Node.js-compatible runtime for edge environments, completing in weeks what traditionally takes months or years. By integrating AI agents into their core …
How Codex opens new possibilities for Notion
Notion integrates OpenAI's Codex across its engineering workflow, treating AI as a teammate rather than an autocomplete tool. Engineers use Codex to one-shot code-aware product specs, ship user-facing…
Scientific computing in the era of AI agents
Science depends on aging, patchwork software that struggles with growing datasets. A new field report shows how AI coding agents—tools that can explore codebases, plan changes, write code, and run tes…
GPT‑6 Astra enables Cognition's Devin to verify its own work
GPT-6 Astra reportedly enhances Devin's core weakness: verification. Instead of just generating code, the AI agent can now run tests and present evidence that software works as expected. This shifts t…
OpenAI teams up with Dell to deliver Codex for hybrid and on-premises enterprise deployments
OpenAI and Dell have announced a partnership to bring the Codex AI coding agent into hybrid and on-premises enterprise environments. The deal lets organizations in regulated industries run Codex withi…