AI for Coding
AI is reshaping software development faster than any other profession — from Copilot autocomplete to agents like Claude Code and Cursor that write, fix and refactor whole projects. "Vibe coding" — building software through natural-language dialogue — has entered the mainstream. Here is everything on AI coding tools: releases, benchmarks (SWE-bench) and the debate over quality and productivity.
Anthropic cuts the data leaking to Claude distillers by 90–99%
Anthropic says new methods cut the data collected by distillation attacks on Claude by 90–99%. The price of that protection: giving up trillions of tokens of…
Anthropic opens its fifth APAC office: Singapore ranks 2nd of 121 countries for Claude
Anthropic opened a Singapore office and signed an MOU with the country's digital ministry. Singapore ranks second of 121 countries in per-capita Claude.ai usage.
Google Turns Gemini Into an Agent With Its Own Email Address and Subagents
Google is turning Gemini into a business agent that plans tasks, delegates to subagents, uses multiple models and gets a work identity with its own email address.
VAST Open-Sources a World Model: Anime Version of Flash Step Runs in Real Time
Claude Mythos 5 Faked Identities to Push Malicious Code: Anthropic Breaks Down 9 of 43 Tests
NX-AI's KronLSTM Changes the Delta Rule and Stabilizes Memory Without Normalization
Ant Shuke Releases LingDT-Decision-1.0: 4B and 27B Models for Fast Decisions
Cognition opens London office as UK team grows 200% in six months
Devin's maker opened an office at The Gilbert on Finsbury Square. Its UK team grew by more than 200% in six months, and Access Group rolled Devin out to 2000…
8B-Parameter Model Beats Claude Opus 4.8 at Generating Scientific Ideas
Salesforce researchers trained an 8B model with set-level reinforcement learning to generate diverse scientific ideas. On a contamination-free benchmark, it…
Vercel launches BYOC on AWS and an agent authorization beta as AI token usage quadruples
Lovable launches Deploy on Azure: your apps and data live in your own subscription
DeepSeek releases V4.1-Flash-1017 and opens its Search API at $0.75 per 1000 calls
Anthropic Hands Agent Skills Standard to AAIF: 1 Million Skills, 280 Million Installs
PaperFormalizer Turns LaTeX Papers into Verified Lean Code, with the Author in the Loop
System Prompt Ablation: How I Shrank My Claude Code Config from 19,736 to 3,585 Lines Without Losing Quality
32.5 Million ₽ a Year for 300 People: How We Quantified the Real Economics of AI Adoption at a Company
RSIGame: a 27B model combo plays its own games and outperforms GPT-5.5
China's RSIGame has AI test, fix and refine games on its own. The authors claim a combination of 27B models beat GPT-5.5 in one-shot generation.
S-Agent 2.0 teaches a coding agent to understand 3D and beats SpatialClaw by 7.8 points
A ready-made coding agent with a set of 3D tools lifted Qwen3.8-27B to 66.5 on MMSI-Bench and 82.3 on ReVSI. Gemini-3.8-Flash reached 74.1 and 90.5.