I am a software engineer and educator with over a decade of experience delivering technical training and building web and cloud solutions. I led training programs and coordinated web development courses at Grupo Impacta — the largest IT training center in Latin America — and trained hundreds of professionals in in-person and remote settings. My current focus is on agentic systems, AI agent orchestration, and scalable cloud-native architectures.
Expertise designing and deploying agentic systems, intelligent agents, and multi-agent ecosystems, including:
- LangChain
- LangGraph (state machines, checkpoints, multi-agent supervisors)
- Google ADK
- CrewAI
- Agent orchestration and multi-agent coordination
- Agentic systems architecture (ReAct, tool use, RAG pipelines, MCP)
- Intelligent automations with n8n, Make, and Zapier
Practical experience crafting prompts and reasoning strategies to improve LLM behavior and performance:
- Zero-Shot and Few-Shot prompting
- Chain-of-Thought (CoT)
- Tree-of-Thoughts (ToT)
- ReAct (Reasoning + Acting)
- Prompt templates, dynamic prompting, and prompt evaluation
Worked with major LLM providers and cloud AI platforms for model access, fine-tuning, and deployment:
- Google Gemini
- Google Vertex AI
- Google ADK
- Microsoft Foundry (formerly Azure AI Foundry)
- AWS Bedrock
- OpenAI
- Anthropic Claude
Strong background building end-to-end web, AI, and cloud solutions:
- Organized and delivered web development training programs for hundreds of students at Grupo Impacta.
- Designed and deployed agentic systems integrating LLMs, multi-agent orchestration, and RAG pipelines for automation and decision support.
- Built cloud-native apps and integrations with CI/CD, observability, and automated deployment pipelines.