docs: 4 posts sobre harness engineering y patrones de orquestación#9
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Posts incluidos
El harness vale más que el modelo — post técnico denso basado en NLAH (Tsinghua) y Meta-Harness (Stanford). El gap del 6×, la ablation surprise, la transferibilidad del harness entre modelos, el arnés mínimo accionable.
Los 10 niveles de complejidad de los agentes IA — taxonomía evolutiva de ReAct hasta Puppeteer/RL. Cada nivel resuelve el límite del anterior. Dónde está la industria y por qué el salto de L5 a L6 es el más difícil.
El patrón correcto para cada problema — 6 dolores reales (agente sin autocorrección, degradación en tareas largas, parallelismo que se pisa, tests que no atrapan bugs, obsolescencia por modelo, supervisión binaria) con su patrón solución y por qué funciona.
Cuatro fuentes independientes, los mismos cuatro principios — Anthropic + OpenAI + Factory + Academia convergiendo en: harness > modelo, separar generación/evaluación, lógica en texto, estado externo. Con la matriz 4×4 y las implicaciones prácticas de cada convergencia.
Ficheros
blog/2026-05-09-harness-engineering-vale-mas-que-el-modelo.mdblog/2026-05-09-diez-niveles-complejidad-agentes-ia.mdblog/2026-05-09-patron-correcto-para-cada-problema.mdblog/2026-05-09-cuatro-fuentes-mismos-principios.mdTest plan
/blogtras el deploy<!-- truncate -->corta bien en el listado en todos los posts