I build systems with language models, agents, knowledge graphs, and automation workflows focused on solving real-world problems.
developer / architect / builder
profile / editorial lab
Juan Iturbe
current focus
Applied AI with product judgment
I work between implementation, operations, and structure: turning AI ideas into clear, observable, usable systems.
Current role
LangGraph Developer & Applied AI Lead
at Ketzai.com Panama City, Panama.
agents LangGraph automation internal systems
Notes
selected context
Currently leading agentic AI implementation at Ketzai: a SaaS application to transform educational planning and delivery.
I focus on solving problems: reducing friction, improving efficiency, designing clear systems that can be operated and scaled.
I come from a technical-creative background as a Technical Artist in international productions (Nickelodeon, HBO), and today I apply that pipeline and systems mindset to software and AI.
operating signal
Python / TypeScript / LangGraph
mode
product / operations / systems
base
Panama City / remote
01
Experiments
Experiments, tools, and applied systems presented like a clear working table: less noise, better reading rhythm, direct access to the details.
01 Agent Systems · R&D Demos
Open Markdown to Knowledge Graph Pipeline
An experimental pipeline that transforms structured markdown documentation into a Neo4j knowledge graph.
Agent Systems · R&D Demos
Markdown to Knowledge Graph Pipeline
An experimental pipeline that transforms structured markdown documentation into a Neo4j knowledge graph.
Agent Systems · R&D Demos
02 Agent Systems · R&D Demos
Open Graph-RAG with Generative UI
Graph-RAG system modeling brand guidelines and assets as a Neo4j knowledge graph for AI agents to traverse and reason over.
Agent Systems · R&D Demos
Graph-RAG with Generative UI
Graph-RAG system modeling brand guidelines and assets as a Neo4j knowledge graph for AI agents to traverse and reason over.
Agent Systems · R&D Demos
03 Workflow Systems · R&D Demos
Open AI Analytics Reporting Workflow
An experimental analytics workflow that extracts Google Analytics data, applies LLM-based analysis, and publishes structured executive reports to Notion.
Workflow Systems · R&D Demos
AI Analytics Reporting Workflow
An experimental analytics workflow that extracts Google Analytics data, applies LLM-based analysis, and publishes structured executive reports to Notion.
Workflow Systems · R&D Demos
04 Agent Systems · R&D Demos
Open Multi-Agent Content Orchestration
An experimental multi-agent workflow built to explore state management, deterministic handoffs, and evaluations in agent systems.
Agent Systems · R&D Demos
Multi-Agent Content Orchestration
An experimental multi-agent workflow built to explore state management, deterministic handoffs, and evaluations in agent systems.
Agent Systems · R&D Demos
02
Notes
An editorial view into ideas, process, and learning around AI, product work, and system-building.
10 min read AI / Architecture / LangGraph / TypeScript
Operations, Not Replacements: CQRS Reducers for Agent State
When your agent's state gets complex, full replacements break. Learn how to use incremental operations with CQRS reducers in LangGraph.
Read article
3 min read AI / Panama / Implementation / Teams
How to Choose an AI Developer in Panama for Your Company
What to look for in an AI developer in Panama: technical judgment, systems thinking, and the ability to ground implementation.
Read article
5 min read AI / Architecture / TS / LangChain
Key Concepts for Implementing Memory in AI Agents
An introductory explanation of the concepts to consider when implementing short-term and long-term memory in agents.
Read article
3 min read AI / Business / Panama / LangChain / Implementation
How to Implement AI Agents in Your Company
Practical guide with local use cases, recommended tools, and implementation strategies specific to the Panamanian market.
Read article
