Tech News
Command and Conquer Generals natively ported to macOS, iPhone, iPad using Fable
A developer used Ghidra + LLM workflow to reverse engineer and port Command & Conquer: Generals to macOS/iOS/iPad, demonstrating LLM-assisted decompilation for game revival.
sqlite-utils 4.0rc2, mostly written by Claude Fable (for about $149.25)
I wrote about the sqlite-utils 4.0rc1 release a couple of weeks ago. Since we only have Claude Fable on our Max subscriptions for a few more days, I decided to see if it could help me get to a 4.0 stable release.
huggingface/transformers Release v5.13.0
Release v5.13.0 New Model additions KimiK 2.5, 2.6, and 2.7 This release includes the architecture for Kimi 2.5 which is used by 2.5-2.7: Kimi K2.5 is an open-source, native multimodal agentic model that advances prac.
Best practices for multi-turn reinforcement learning in Amazon SageMaker AI
In this post, we share best practices for reliable multi-turn RL training. We cover how to build a training environment you can trust, set up an external evaluation, design a reward aligned with the end task, manage.
Astrophysicists Puzzle over Webb’s New Universe
Astrophysicists are puzzled by new James Webb Space Telescope observations of 'little red dots' that may be black holes cocooned in gas or brown dwarfs, challenging existing models.
GitHub Repos
Panniantong/Agent-Reach
Give your AI agent eyes to see the entire internet.
# Python# agent infrastructure# ai agentzilliztech/claude-context
Code search MCP for Claude Code.
# TypeScript# agent# agentic ragmksglu/context-mode
Context window optimization for AI coding agents.
# TypeScript# antigravity# claudeResearch Papers
ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning
ReContext improves LLM long-context reasoning by recursively selecting and replaying relevant evidence from the input without training or external memory.
When LLMs Develop Languages: Symbolic Communication for Efficient Multi-Agent Reasoning
This paper introduces a method where multiple LLM agents create and use compact symbolic languages to communicate, reducing token usage by 3-6x while maintaining reasoning accuracy.
SPLC: Social Preference Learning for Crowd Robot Navigation
A Social Preference Learning for Crowd Robot Navigation (SPLC) algorithm to eliminate the need for detailed reward design and introduce a social preference feedback mechanism to automatically generate preference data through principled preference evaluation criteria.