Criador do bridgecode

Marcelo Santiago

Developer and creator of bridgecode, an agentic development environment for orchestrating squads of AI agents. Helps people going from vibe coding to professional development move from idea to shipped product.

GitHub Twitter / X LinkedIn Instagram Threads Website

Artigos publicados

252 artigos por Marcelo Santiago

Benchmarks

Oct 25, 20265 min

Nanbeige4.3-33B: agent benchmark results

The card updated on October 25 compares four models on agent tests, with results from the Nanbeige team itself and limitations in the evaluation.

#Nanbeige#Nanbeige4.3-33B#benchmarks

Marcelo Santiago Read

Official Arena chart highlights Kimi K3.1 Pro (Max) in 11th place on Agent Arena, with a net improvement of 7.6% over baseline.

Benchmarks

Oct 25, 20263 min

Kimi K3.1 Pro debuts on Agent Arena in 11th place

Arena added Kimi K3.1 Pro to Agent Arena on October 24. The chart shows the model in 11th place, with +7.6% versus baseline.

#Kimi K3.1 Pro#Agent Arena#Arena.ai

Marcelo Santiago Read

Community article card for AtMem 2.3.10 on Hugging Face, with the title about evidence discovery for persistent agent memory and the author's name

Newsletter

Oct 24, 20265 min

AtMem 2.3.10: read-only evidence discovery for agent memory

AtMem 2.3.10 separates evidence discovery from context injection. The reported DolphinBench result (390/600) and what the evaluation does not prove.

#AtMem#memória de agentes#DolphinBench

Marcelo Santiago Read

Official release card for Qwen Code v0.26.1 on GitHub, with the titles Highlights and Breaking Changes

Newsletter

Oct 24, 20264 min

Qwen Code 0.26.1: save the daemon token with --save-token

Qwen Code 0.26.1 adds qwen daemon start --save-token, context compression for idle sessions and session isolation fixes. Here is how to use them.

#Qwen#Qwen Code#release

Marcelo Santiago Read

Claude Code v2.1.310 release card with the official list of changes for Monitor and plugin monitors

Newsletter

Oct 24, 20264 min

Claude Code 2.1.310: new Monitor tool for Bedrock, Vertex and Foundry

What's new in Claude Code 2.1.310: the Monitor tool, plugin monitors on Bedrock, Vertex and Foundry, and the session and permission fixes to know about.

#Anthropic#Claude Code#release

Marcelo Santiago Read

Official image of OpenAI Daybreak Patch the Planet with the project name and an ASCII character globe, linked to the Brotli vulnerability CVE-2026-111497 notice.

Newsletter

Oct 24, 20264 min

Brotli RCE (CVE-2026-111497): OpenAI flaw hits Node.js, Bun and Envoy

OpenAI found unauthenticated remote code execution in the Brotli decoder. See which software ships it (Node.js, Bun, Envoy), how to check exposure and how to patch.

#OpenAI#Brotli#segurança

Marcelo Santiago Read

Official GitHub Copilot illustration about the metrics API, with text indicating the inclusion of requests and tokens from tools like OpenCode.

Newsletter

Oct 24, 20264 min

Copilot API includes usage via OpenCode

The GitHub Copilot API now breaks down requests and tokens by client, including OpenCode, in daily and 28 day reports.

#GitHub Copilot#OpenCode#API

Marcelo Santiago Read

Anthropic's official cover for the article on security, privacy and protections for the Claude Duet agent.

Newsletter

Oct 24, 20264 min

Claude Duet: Anthropic details security and privacy

Anthropic explained how Claude Duet combines approvals, external checks beyond the model and isolated credentials before acting on connected services.

#Anthropic#Claude Duet#segurança de agentes

Marcelo Santiago Read