DEV Community

Team Quesma profile picture

Team Quesma

Independent evaluation and training for the AI agent ecosystem. Real-world complexity through simulation environments where agents face multi-hour tasks.

Joined Joined on  Personal website https://quesma.com/
Baba Is Solved by Fable 5 and GPT-5.6 Sol, but at what cost?
teamquesma profile

Baba Is Solved by Fable 5 and GPT-5.6 Sol, but at what cost?

#ai #quesma #gamedev
Comments
10 min read
Tokenflation: When “Hi” triggers 33 tool calls
teamquesma profile

Tokenflation: When “Hi” triggers 33 tool calls

#ai #llm #programming #quesma
Comments
5 min read
Qwen3.6 27B is the sweet spot for local development
teamquesma profile

Qwen3.6 27B is the sweet spot for local development

#ai #llm #quesma
Comments
7 min read
Compare harnesses not models: Blitzy vs GPT-5.4 on SWE-Bench Pro
teamquesma profile

Compare harnesses not models: Blitzy vs GPT-5.4 on SWE-Bench Pro

#ai #llm #openai #harnessengineering
Comments
7 min read
We hid backdoors in binaries — Opus 4.6 found 49% of them
teamquesma profile

We hid backdoors in binaries — Opus 4.6 found 49% of them

#ai #benchmark #security
Comments
13 min read
Reverse engineering River Raid with Claude, Ghidra, and MCP
teamquesma profile

Reverse engineering River Raid with Claude, Ghidra, and MCP

#claude #ai #ghidra #mcp
1
Comments
4 min read
Vibe coding needs git blame
teamquesma profile

Vibe coding needs git blame

#ai #claude #vibecoding
Comments 1
7 min read
How 2025 took AI from party tricks to production tools
teamquesma profile

How 2025 took AI from party tricks to production tools

#ai #llm
Comments
6 min read
Antigravity feels heavy and Claude Skills are light
teamquesma profile

Antigravity feels heavy and Claude Skills are light

#ai #claude #antigravity
1
Comments
7 min read
loading...