Tag
8 articles
Learn to build a code agent evaluation system that detects reward hacking in benchmarking, where agents retrieve known fixes instead of deriving solutions.
A new Anthropic study reveals that men use AI coding agents more than twice as often as women in social science research, highlighting a significant gender gap in AI tool adoption.
Warp integrates OpenAI's GPT-5.5 model to enhance coding agents across local, cloud, and open-source development workflows. The move positions Warp as a key player in AI-powered development tools.
Learn how to set up a benchmarking framework to evaluate AI coding agents like Claude Code and GPT-5.5, similar to industry benchmarks used in 2026.
OpenAI details its comprehensive security approach for running Codex, including sandboxing, network policies, and agent-native telemetry to support safe and compliant AI coding agent adoption.
Alibaba's Qwen team has released Qwen3.6-27B, a dense open-weight model outperforming 397B MoE on agentic coding benchmarks. It introduces a Thinking Preservation mechanism and a hybrid attention architecture.
Learn how to install and use Context Hub, an open-source tool that keeps coding agents updated with the latest API documentation from Andrew Ng's team at DeepLearning.AI.
Learn how too much detail in AI coding instructions can actually hurt performance, according to a new ETH Zurich study. Understand the concept of context engineering and why less can sometimes be more when guiding AI systems.
