Workflows vs. Agents: Choosing the Simplest Thing That Works
Most LLM features don't need an agent. A practical way to decide between a single call, a workflow, and a fully autonomous agent.
Open to opportunities
Rolando Arroyos · Software Engineering & AI
Lead Software Engineer | Generative AI | Agentic Systems
Exploring the intersection of enterprise software engineering and artificial intelligence. From scalable architectures to autonomous AI agents, I turn emerging technologies into practical solutions—and share what I learn along the way.
Areas of focus
Distributed systems, API strategy, and modernization plans built for resilient delivery.
Practical applications of generative AI, retrieval, and agentic workflows with thoughtful guardrails.
Cloud-native architectures and integration patterns that connect teams, data, and services.
Clarity across complex programs—from design decisions to the teams that carry them through.
From the blog
Most LLM features don't need an agent. A practical way to decide between a single call, a workflow, and a fully autonomous agent.
An agent is only as good as its tools. Naming, schemas, error messages, and output size matter more than the framework you pick.
The pieces around the model loop — state, sandboxing, observability, evals, and cost controls — that turn a demo into a dependable system.
Keep in touch
Get practical notes on AI systems, enterprise architecture, and engineering leadership.
Newsletter
Practical notes on AI agents and system architecture. No spam, unsubscribe anytime.