#编码

With AI writing the code, entrepreneurial thinking is now what counts most for newcomers at Anthropic

Anthropic engineering manager Felix Rieseberg has scrapped the classic coding interview, arguing Claude now handles small programming tasks better than candidates and that "hunger beats tenure" as newer engineers overtake experienced colleagues. He advises developers to learn their manager's manager's goals and build a weekend prototype of one ambitious idea, citing two relatively new Anthropic engineers who built the Claude Code graphical interface over a weekend before a public release about eight days later. The article notes his advice shifts entry-level hiring toward entrepreneurial thinking, product sense and taste, skills most universities barely teach.

0.166.0-alpha.5

The openai/codex repository published the rust-v0.166.0-alpha.5 release tag, tagged by fcoury-oai on 28 Oct at 02:58. The release page lists two assets.

Announcing Artificial Analysis Intelligence Index v5

Artificial Analysis released Intelligence Index v5, its biggest upgrade since v4, adding FrontierCode 2.0 and Terminal-Bench-Science and upgrading HLE-Diamond, AA-Omniscience v1.1 and CritPt v2, with 60% of the Index now based on private test sets or held-out answers. Claude Opus 5.5 (max) leads at 58, ahead of GPT-6 Astra (max, 57) and GPT-6.1 Sol (max, 55), while the open weights gap widens from 3 points to 12. It also launched the Progress Index v1, which tracks model capability over time on one continuous scale.

Do Guardrails Make AI Coding Agents Better? What 270 Runs Showed

An experiment of 270 unattended coding-agent runs across three models and seven tasks found that instruction files and pre-commit gates did not raise first-pass rates, with 267 of 270 runs passing hidden tests at first "done." Guardrails showed up instead in what runs left behind: on four later tasks, 9 of 60 gated runs left a defect versus 15 of 60 ungated, a gap too small to call. The author notes limits including five runs per cell, one small codebase, and self-written tasks and harness, and publishes the run records and a reproduction script in a companion repository.

0.166.0-alpha.4

OpenAI's Codex repository published the rust-v0.166.0-alpha.4 release tag, tagged by fcoury-oai on 27 Oct at 23:31. The release page lists two assets.

OpenCode-Muse: Use Meta AI's Muse Code Subscription in OpenCode

An OpenCode plugin that lets OpenCode and OpenChamber use a Muse Code subscription while authentication and inference stay inside the official Muse Code CLI (muse serve, via @muse-code/sdk), with no subscription credential extracted, cached, or injected into OpenCode. Users add the published package to their opencode.jsonc, sign in with muse login, restart, and pick a Muse model; billing defaults to the Contributor lane with an opt-out to Standard via the OPENCODE_MUSE_CONTRIBUTOR_MODE environment variable. The plugin supports text, image, video, audio (transcribed via muse voice transcribe), and PDF attachments, per-model reasoning effort, conversational answers to Muse permission prompts and option-questions, and per-session worktrees where the CLI supports them.

Why Programming Is Especially Exposed to AI Automation

The author argues that programming is unusually exposed to AI automation for six reasons: it is intellectual rather than physical work, involves little direct client contact, has few formal barriers to entry, produces results that can be verified automatically through testing, separates responsibility from implementation, and requires no human face. He connects this to Fred Brooks's "no silver bullet" argument, noting that AI can now handle the essential complexity of turning imprecise requirements into code. He also observes that some young people are already avoiding IT careers because they doubt their future-proofing.

Show HN: Make your AI agent think before it codes

SpecOps is an open-source tool that adds a structured four-phase workflow to AI coding assistants: understanding the codebase, writing requirements and design, implementing ordered tasks from the spec, and verifying acceptance criteria. It works with Claude Code, Cursor, OpenAI Codex, GitHub Copilot, and Google Antigravity, storing git-tracked markdown specs, and adds features like EARS-notation requirements, drift checks, dependency gates, adversarial evaluation, and a production learnings layer.

0.166.0-alpha.3

The openai/codex repository published the release tag rust-v0.166.0-alpha.3 (commit 4f7bfec), tagged by jif-oai on 27 Oct at 17:33. The release page lists two assets.

Python Typing Survey 2026: AI Tools and the Shifting Checker Landscape

The 2026 Python Typing Survey, with 1,794 responses, found type hint use stable across three years (85.5% in 2026) while checker selection shifted sharply, with 49.9% of respondents selecting two or more checkers and speed cited by 56.5% as the top reason for choosing one. Missing type information in third-party libraries was the leading difficulty at 65.1%, and AI tools wrote type hints for 42.5% of 1,517 respondents, with 24.4% saying an AI coding tool invoked their checker.

AI review optimized for recall (catches more defects than Codex or Claude)

QuickReview is a self-hosted GitHub PR reviewer that fans each change out to many narrow, read-only lens passes, merges duplicate findings, and re-validates every candidate against the code before posting one review. On 185 real PRs with 520 verified defects it finds 97% of high-severity, 93% of medium and 84% of low defects, versus 90%, 82% and 52% for the best of six other reviewers, at about $2 and 17.5 minutes per PR. It is in beta, runs as a deterministic TypeScript daemon with no LLM in its decision path, and ships with dry_run = true so a fresh install posts nothing.

Show HN: Snippet – An Always-On AI Agent

Snippet is a coding agent team that runs on your own computer and can be prompted from a phone, desktop, or terminal, with a Mission Control layer that assigns work to agents, checks results, and returns finished changes. It runs on the model you already pay for, moving background work to the next model when one hits its limit, and each agent has its own memory stored as plain files in your repo. It supports scheduled jobs, quiet hours, and sessions shared between the terminal app and phone, and installs via a curl script from snippet.wacht.dev.

Coding From the Himalayas With an Offline AI Agent

A Thomson Reuters principal engineer describes running the OpenCode coding agent with Ollama and Qwen3.5 9B entirely on a laptop, then disconnecting the internet to fix a bug in an expense calculator and run its tests. The article covers model size and memory considerations, published benchmark scores, setup steps, and the limits of what can be done offline.

Mirendil Launches Fern 0.1, Its First Model, With 10M-Token Context

Mirendil released Fern 0.1, a from-scratch coding model with a context window of up to 10M tokens (about 23 MB of code), built on Mirendil-Base after a short round of post-training and available only through hosted early access. Mirendil says one researcher pretrained the base model in 3.5 months for under $10M, with AI agents writing over 99% of the code and training data drawn only from public sources. The base model scores 100% on RULER single-needle retrieval from 32K to 10M tokens, while Fern 0.1 scores 6.1% on Terminal Bench 4.0 and 73.4% on SWE-Bench Verified, trailing the frontier and large open models in its comparison table.

Show HN: Give our AI coding runner a real task and watch it until merge

mcp4mail.online is a live page where visitors can pick a real task for an open-source mail project, confirm by e-mail, and follow the AI runner from branch to pull request, tests and merge on GitHub. The site reports 179 finished tasks in 40 days, with runners on the team's own machines, no direct main-branch access, and work hours of 8:00–18:00 Prague time on work days. The page lists current tasks such as GitHub issue templates, a self-hosting docs dry run, a README provider table and screenshots, and a mail-index unread-state fix.

0.166.0-alpha.2

A pre-release tagged rust-v0.166.0-alpha.2 was published by github-actions on 27 Oct at 11:54, with 4 commits to main since the release. The release page lists 288 assets.

Ask HN: Can we do something about this impending AI dystopia

An Ask HN post argues that developers are becoming increasingly dependent on agentic coding, which is accelerating the end of the software profession, and asks the community how to slow or halt this trend. The author compares developers to a drug addict who knows the drug may end him yet keeps consuming it, and links to two prior HN threads on the topic.

0.165.0-alpha.1.2

A pre-release tagged rust-v0.165.0-alpha.1.2 was published by github-actions on 27 Oct at 04:57, with 151 commits to main since the previous release. The release is identified by commit 720fec8 and lists 320 assets.

0.165.0-alpha.1.1

A pre-release tagged rust-v0.165.0-alpha.1.1 was published by github-actions on 27 Oct at 00:36, with 151 commits to main since the previous release. The release page lists 288 assets.

GPT-6.1 Sol helps CodeRabbit strengthen security across complex codebases

CodeRabbit uses GPT-6.1 Sol to investigate security risks across complex codebases. Its security agent helps developers review findings and proposed fixes.

v2.1.311

Claude Code v2.1.311 adds --hold-idle-file and --max-hold-min to self-hosted-runner, fixes startup freezes with long CLAUDE.md list items, and resolves standard-context sessions stopping with "Usage credits required for 1M context" by compacting and continuing. It also fixes Write/Edit stalls on large files, subagent MCP tool access, stale MEMORY.md/CLAUDE.md after compaction, and numerous MCP, rewind, IDE, and cloud session issues, plus changes ultrathink to run at xhigh effort on Opus 5.5, Sonnet 5.5, Haiku 5.5 and Fable 5.1 via Anthropic's API.

GPT-6.1 Terra halves OpenAI's API prices and still gains ground on its larger models

OpenAI released GPT-6.1 Terra, priced at $1 per million input tokens and $5 per million output tokens—half the cost of GPT-5.6 Terra and below Claude Sonnet 5.5's $2/$10. Artificial Analysis reports it gains 6 Intelligence Index points over GPT-5.6 Terra at maximum reasoning effort while costing $0.40 versus $1.40 per task, and it solves 50.3% of Terminal-Bench 4.0 tasks versus 21.5% for its predecessor. It is available now for Plus, Pro, Enterprise, and Edu users in ChatGPT Work and Codex and via the API, though OpenAI still recommends GPT-6.1 Sol and GPT-6 Astra for the most demanding coding tasks.

Featured coverage of this story: 《Introducing GPT-6.1 Terra》

How Ultrafast cut incident investigation from hours to minutes at OpenAI

OpenAI describes how its engineers use Codex Ultrafast to investigate incidents in minutes rather than hours. The post also covers faster security analysis, code refactoring, and daily workflows.

Can AI help solve the COBOL code crisis?

OpenAI reported that its latest models, including GPT-6.1 Sol, showed "substantial improvements" in migrating legacy code such as COBOL into modern languages like C++, Java, Kotlin, Python, and Rust while preserving behavior, with GPT-6.1 Sol reaching 87 percent behavior preservation accuracy at low reasoning effort and 98 percent at higher reasoning. Experts told The Verge that such benchmarks are far from enabling safe replacement of critical legacy systems, stressing that rewriting code is not the same as replacing a system and that breaking complex systems into replaceable units remains the key difficulty. OpenAI spokesperson Oscar Haines acknowledged legacy code "is still far from a solved problem."

Kevin Roose is AGI-pilled. Why aren’t you?

In this Decoder interview, Kevin Roose discusses leaving The New York Times with Casey Newton to launch the video-first podcast Machine Gods, and his new book The AGI Chronicles, a reported history of the frontier labs racing to build artificial general intelligence. He describes how the transformer gave companies a tool for a long-held dream, how labs differ in how "AGI-pilled" they are (with Anthropic most and Google least), and how belief in AGI drives both bet-the-farm compute spending and a struggle to focus on business fundamentals.

battledrones v14: Compass Rose, a four player map

The developer added a four-player online mode to Battledrones, whose previous maps were limited to two players, using a prompt asking for a new map with four-way rotational symmetry, up to four AI players, and support for the WebSocket non-audio multiplayer mode so up to four humans can join with spare slots filled by AI. Claude Opus 4.5 added a map called "Compass Rose" and modified the WebSockets multiplayer network code to support up to four players for that map, joining via share links or QR codes. The new teams use purple and green robot themes alongside the existing blue and red designs, previewable in the updated model browser.

Fancy harnesses make coding agents more expensive but not better, study finds

A study by Liu and Han ran five models through Claude Code, OpenCode and mini-SWE-agent and found that swapping harnesses shifts SWE-bench Verified results about as much as simply rerunning the same setup, with differences of two to five solved tasks per model. The only measurable harness effect was a deficit for OpenCode, while Claude Code's roughly 16,600-token system prompt and tool descriptions made a single task cost up to three times more than the leanest harness. The authors conclude a harness buys fewer ways to lose a task and a bill, not pass rate, and recommend publishing a rerun of the same setup alongside every harness comparison.

[AINews] Kimi 3.1 Pro 104BA-2.8T: Fable-5.1-Class Open Weights narrows Closed-AI lead to 1 month... and Pro Highspeed API with 2-3× speed

Moonshot launched Kimi K3.1 Pro, a multimodal 3T-class mixture-of-experts model with 2.8 trillion total parameters, 104 billion active parameters and a 1-million-token context window, available via API with weights promised for November 4, 2026. Artificial Analysis placed it fourth on its Intelligence Index at 55, behind Opus 5.5 at 58 and Sonnet 5.5 and Muse Wonder Full at 56, and said the leading-open-versus-proprietary gap would narrow from 12 points to three once weights are released. Standard pricing is $2/M input, $10/M output and $0.20/M cached input tokens, with a Highspeed version at twice the price that Moonshot claims generates output 2–3× faster.

Quick Study

A prompt to Claude Opus 5.5 High asks it to build an artifact that opens a photo and, within a few seconds, paints a version of that photo using a canvas element and a simulated paintbrush. The post also references a previous experiment in which Claude Fable simulated a watercolor illustration with P5.js.

Chinese AI model Kimi K3.1 Pro is closing in on Anthropic's Claude Opus 5.5

Moonshot AI released Kimi K3.1 Pro, a post-trained successor to Kimi K3 that scores 55 points on the Artificial Analysis Intelligence Index, an 11-point gain that places it fourth behind Claude Opus 5.5 (58), Claude Sonnet 5.5 (56) and Muse Wonder Full (56). Moonshot says the model keeps K3's 2.8 trillion total and 104 billion active parameters with a one-million-token context window, cuts token prices by 33 percent, and shows its largest gains on agentic tasks, while cyber-related training data was stripped for the planned open-weights release. A faster K3.1 Pro Highspeed variant is also available through Kimi.ai, Kimi Work, Kimi Code and the Kimi API.

Anthropic's Claude now writes code and proves it safe automatically

Anthropic's Frontier Red Team, working with partner Theorem, used Claude to generate roughly 180 million lines of mathematical proof code in a few weeks, verifying existing software from 123 open-source libraries in Google's OSS-Fuzz project by translating crash tests into mathematical problems and analyzing compiled machine code, with the Rocq program automatically checking the proofs. The researchers say Claude found hundreds of bugs, including as many security bugs in one week in a widely used PDF reader as are typically reported for that software in a year, and caught a logical error undetected for over a decade. Anthropic notes the proofs hold only under strict conditions—a specific processor architecture, single-threaded programs, and inputs under 64 MB—and that compute costs are steep, with individual proofs running up hundreds of millions of tokens.

Recursive launches Flint, a laptop coding model it says was largely built by AI

Recursive released Flint, a free coding model that runs on a laptop and was built largely with the help of its automated research system, the Eureka Machine, letting a small team build it from scratch in under four months. Flint is a mixture-of-experts model using 2.5 billion of its 40 billion parameters at a time, is available on Hugging Face under the Apache 2.0 license, and in Recursive's tests generated about 131 tokens per second on a MacBook Pro with an M4 Pro chip. The company says the longer-term goal is for the Eureka Machine to achieve full recursive self-improvement, with humans overseeing the process.

Recursive shows first result of its AI-driven research process with Flint

AI startup Recursive has released Flint, a mixture-of-experts model with 40 billion total and 2.5 billion active parameters, built in four months with the help of its "Eureka Machine" system that proposed, ran, and evaluated experiments. Recursive claims Flint leads coding benchmarks among MoE models under 50 billion parameters, though the results have not been independently verified, and the weights are available on Hugging Face under Apache 2.0. The company says Flint runs locally with context windows up to 512,000 tokens and generates over 125 tokens per second on an M4 Pro, but acknowledges it needs more guidance on complex tasks than large proprietary models.

Battle Drones - a vibe coded RTS with audio-based multiplayer

Simon Willison describes Battle Drones, a browser-based Command & Conquer-style RTS built largely by coding agents such as Claude Code over 12 major iterations and 321 commits, with 3D models, fog of war, multiple maps and a random map generator, several game modes, and AI personalities. It supports multiplayer between two nearby devices by synchronizing game state over audio, and an online multiplayer mode was later added via a stateless WebSockets relay server on Fly.io. The post also covers agent-vs-agent battles where two LLMs play each other, and lessons such as using a deterministic simulation engine for replays and multiplayer.

How 3 Java Champions are using GitHub Copilot for increased productivity

GitHub interviews three Java Champions—Vlad Mihalcea, Patrick Baumgartner, and Frank Delporte—about using GitHub Copilot in production and open source work, including a Hypersistence Optimizer app built in two days, a 900-slide Reveal.js to Marp migration, and diagnosing a Pi4J bug where virtual threads pinned in native poll() calls left a fifth button unresponsive. They emphasize reviewing generated code, defining verification gates, and keeping humans accountable for what ships.

My obituary for programming

Jeremy Scheff reflects on what career programmers lose as coding agents take over: programming was once a job where you were paid to solve tricky logic puzzles, and the harder the puzzle the more you were paid, because regular people couldn't do it and needed programmers to set strategy, timelines and what was possible. He notes that programmers who thrive on building working systems and don't care deeply about the problems solved along the way benefit greatly from LLMs, while those who enjoyed directly solving those problems face a very real loss.

How to continue a GitHub Copilot app session in VS Code

This GitHub tutorial explains how to open a GitHub Copilot app session in VS Code so its conversation, diff, and worktree carry over, letting you review changes line by line, edit by hand, run tests locally, and push updates to the session's pull request. It lists prerequisites and walks through each step, including prompting Copilot further from the sidebar and committing with normal Git commands.

Introducing Mistral Large 4

Mistral launched a public preview of Mistral Large 4, a 1-trillion-parameter natively multimodal model with 52 billion active parameters, trained from scratch on 3,800 NVIDIA Grace Blackwell GPUs in Mistral's own European datacenters, with weights to be released by the end of the month. Mistral reports competitive performance across coding, agentic workflows, multimodal understanding, cybersecurity, science and knowledge work, claiming it outperforms any open-weight model developed in the US or Europe. The preview API is available on Mistral Studio, and the model will serve as the foundation for a new generation of specialized Mistral models.

ReviewBench: An open benchmark for AI code review

GitHub introduces ReviewBench, an open benchmark for AI code review built from analysis of 103.9 million GitHub pull requests, containing 219 public pull requests across 19 languages with a multi-source golden set labeled for severity and category. It reports grounded and augmented precision, recall, and F1 metrics, uses Claude Sonnet 5 as the LLM grader under a published rubric, and had its ground-truth findings independently re-labeled by senior engineers with 96.6% agreement. GitHub says ReviewBench's offline predictions for a Copilot code review multi-model ensemble matched a later A/B test, which showed an 8.0% rise in addressed rate, 13.6% in recall, 61% in comment volume, and an 8.0% drop in cost per review.

AI is changing developer work. Here are three skills to strengthen.

GitHub outlines three skills developers should strengthen as AI reshapes their work: learning to direct AI agents rather than just use them, not trusting AI's first answer and instead having a second model critique it before review, and using time saved on implementation to tackle broader problems like customer needs and architectural tradeoffs. The post illustrates these with examples such as coordinating multiple agents on an authentication flow and a second model critiquing a SQL query. It notes GitHub Copilot's built-in Rubber Duck agent uses a second model to critique plans, code, and tests.