Construyendo con sistemas de IA

• Choose an architecture based on task and evidence needs.
• Evaluate quality, latency, cost, and safety together.
• Design monitoring and incident response before launch.

Recommended first: Fundamentos de la IA, Usuario responsable de IA

1. 1

   Language-model systems

   Outcome: Reason about tokens, context, generation, and model tradeoffs.

   Practice: Compare two model options using quality, latency, context, privacy, and cost.

   Mecánica del modelo de lenguajeModels, inference, and limitationsModel cost and operational tradeoffs
   ChatGPT y LLM →Modelos de IA explicados →Inferencia de IA →
2. 2

   Grounding with retrieval

   Outcome: Know when and how retrieval can improve evidence access.

   Practice: Design a retrieval test set with answer and citation requirements.

   Generación de recuperación aumentadaVerificación de fuente y reclamo
   Calidad de recuperación →Bases de datos vectoriales →Incrustaciones →
3. 3

   Agents and tools

   Outcome: Bound multi-step systems with permissions and checkpoints.

   Practice: Write an agent permission model and failure-recovery path.

   Agents, tools, and long-running tasksAutomation boundaries and safeguardsAI security and misuse risk
   Agentes de IA →Automatización del flujo de trabajo de IA →
4. 4

   Evaluation and operations

   Outcome: Measure system performance before and after deployment.

   Practice: Create an evaluation suite covering quality, refusal, latency, cost, and regressions.

   Evaluation and benchmark literacySuccess metrics and monitoringAI incident responseExperiment and pilot design
   Puntos de referencia de IA →Seguridad de la IA →