Tech News
← Home  ·  All topics

Code Generation

10 GoKawiil briefs on this topic · RSS for code generation

Microsoft details internal AI overhaul led by CTO Kevin Scott and AI CEO Mustafa Suleyman

Microsoft executives including CTO Kevin Scott and AI chief Mustafa Suleyman described changes to how the company builds software, citing an internal accelerator program, AI-assisted code generation, and custom models and agents now used across engineering teams. The account outlines a shift in Microsoft's development practices driven by its own AI tools.

Essay argues agentic coding tools are turning code into malleable 'clay'

In a Sep 25, 2026 post titled 'Code is clay now,' writer Alberto Denia argues that agentic coding tools have drastically lowered the cost of changing code, making it behave more like clay than stone. He describes this as enabling a 'modeling' approach to building software, where developers sketch rough versions and iteratively reshape them rather than planning meticulously upfront.

New open-source benchmark Parsimony measures code bloat added by AI coding models

A developer has released Parsimony, an open-source benchmark that measures how much code AI models add when solving SWE-bench Verified tasks, rather than just whether they solve them. It parses patches to count syntax-level coding units added and removed, producing a net score, and ranks models by mean net units added on tasks they solved. The tool and underlying data are published on GitHub, with results also shown via an interactive leaderboard and charts.

AI researcher Andrej Karpathy says awareness of LLM capability is splitting into four tiers

Andrej Karpathy argues that public understanding of large language models' capabilities is diverging sharply rather than converging. He outlines a rough tier system: roughly 75% of people have barely used LLMs, 20% use free chatbots casually, about 0.2% of professionals in coding and math see agents completing weeks of work in a prompt, and an estimated 5,000 people with internal access to frontier systems are reportedly seeing swarms of agents handle multi-week software projects, cyberattacks and research-level math.

Montana State professor Carson Gross argues students must still learn to code despite AI

Carson Gross, a computer science professor at Montana State University, says he tells students that even though AI can generate code for assignments, they should write it themselves. He argues programming is fundamentally about problem-solving and managing complexity, skills he believes remain valuable despite AI tools, while warning that junior programmers who rely on AI-generated code risk never developing the ability to read and understand code.

Engineers admit they can no longer fully understand AI-generated code

Industry sources describe software engineers increasingly unable to interpret code produced by large language models, with their jobs shifting from writing software to overseeing AI output. Infinity founder Jeremy Nixon and SemiAnalysis' Jordan Nanos cited examples, including OpenAI engineers reportedly unable to explain a GPU kernel generated for DeepSeek, despite the AI's code passing tests and performing correctly.

DHH says he's stepped back from hand-coding, embraces LLMs and Rust at 37signals

At Rails World 2026, David Heinemeier Hansson said he has retired as a professional programmer, calling himself a 'maker' who directs LLMs to write code in English rather than reviewing the output line by line. He said 37signals' next version of Hey will use LLM-generated Rust on the backend and native apps on each platform instead of the Rails-based web stack he has long championed, and claimed to have produced roughly 150,000 lines of code in August versus about 30,000 lines a year previously.

Essay likens AI coding shift to stonemasons-to-carpenters bridge-building change

An essay draws a parallel between the historical shift in bridge construction from stonemasons hand-placing stone to carpenters building wooden forms for poured concrete, and today's shift in software development driven by AI code generation. The author argues developers are moving from directly writing code line-by-line, like stonemasons, to constructing structures such as tests, documentation and guardrails that shape what AI generates, akin to carpenters building molds.

Developer Details Six Months Coding Only With AI Agents, Not by Hand

A software engineer describes a personal experiment begun in February where they stopped writing code manually and instead relied entirely on AI coding agents like Copilot, Cursor, and Claude Code to implement changes. The account traces how tooling evolved from simple autocomplete suggestions to agents capable of editing multiple files from a single natural-language description, sharply reducing the amount of typing required.

Analysis: AI coding agents are compressing traditional software development workflows

A new industry analysis argues that as AI coding agents rapidly improve and handle longer tasks asynchronously, the traditional sequential product development pipeline—idea, product, design, engineering, QA, production—is being restructured. Implementation, once the dominant time cost in this pipeline, is shrinking, shifting the bottleneck to verification and simulation instead.