AI Briefing#software-engineering
The latest AI and developer news about #software-engineering, with the original source and a short summary.
The latest AI and developer news about #software-engineering, with the original source and a short summary.
A five-step infrastructure design for building an AI coding agent-based software factory, analyzing measured results and limitations from major companies
The SlopCodeBench study found that code generated by AI agents has approximately twice the duplication and complexity of human code, with even the latest models achieving a 0% pass rate on strict tests.
Marc Andreessen stated that Cognition's Devin handles 90% of code writing and boosts engineer productivity by more than 10 times.
OpenAI researchers achieved 3.14 days' worth of work via AI agents, but inference costs surged 40x in five months.
Incomplete code generated by AI appears to work on the surface but hides fatal internal defects, ultimately requiring a rewrite to salvage.
It raises concerns that the radical advancement of AI technology could lead to a loss of humanity.
A Brazilian research team has released 'Reversa,' a multi-agent framework that converts legacy code into operational specifications understandable by AI agents.
As AI lowers coding costs, engineer value derives from judgment based on codebase understanding and system context.
Introduced a Harness Engineering framework combining context management, architectural constraints, and regular cleanup to prevent the degradation of AI-generated code quality.
Anthropic has formalized agent-based development workflows, signaling a shift in the code review paradigm.
GitLab analyzes that as AI makes code production cheap, reliability and governance become the core focus.
Anthropic's new playbook outlines a six-stage AI-native SDLC where Claude Code and agents automate planning, coding, and testing, shifting human roles to orchestration and governance.
The explosion in AI-generated code volume has created code review bottlenecks, leading to a redefinition of roles between humans and AI.
Z.ai's **GLM 5.3** has been added to Vercel's **AI Gateway**, enhancing performance in software engineering and agentic tasks.
Andrew Ng and DeepLearning.AI introduce four core competencies for AI engineers, derived from an analysis of hiring data.
A new development trend in Linux 7.2, which utilizes AI tools to detect and patch security vulnerabilities, has been revealed.
Airbnb CEO Brian Chesky stated that AI is writing 60% of new code, accelerating product development.
Andrew Ng analyzed over 10,000 job postings to identify four core skills for AI engineering.
Antigravity has introduced **Custom Agents**, optimized for specific roles and tools, in version 2.0 and the CLI.
Introduces the methodology of **Harness Engineering**, which designs surrounding environments such as context and tools rather than the model itself to enhance the performance of coding agents.
This case study analyzes the construction of an automated harness that verifies a 55,000-line FPS game generated from a single prompt.
Google has released Gemini 3.7 Flash, a high-performance model optimized for coding and agentic tasks.
The full schedule for GitHub Universe 2026 has been released, focusing on the future of AI-driven development.
Automated security patching and backporting processes leveraging AI agents significantly reduced vulnerabilities in container images.