https://www.fernandohermida.com/tags/ai-agents/Recent content in Ai-Agents on Fernando HermidaHugoen-usMon, 10 Aug 2026 00:00:00 +0000https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/01-hello-world-langchain/Sat, 25 Jul 2026 00:00:00 +0000https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/01-hello-world-langchain/A first LangChain chain, from prompt template to LCEL pipe, with a local model swap, response inspection, tests, type-checking, and tracing added around it.https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/02-agent-loops-in-python/Mon, 10 Aug 2026 00:00:00 +0000https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/02-agent-loops-in-python/What actually happens between an LLM call and a tool call: the read-act-observe loop that turns a plain language model into an agent.https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/03-tool-calling-with-langchain/Mon, 10 Aug 2026 00:00:00 +0000https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/03-tool-calling-with-langchain/How create_agent and the @tool decorator collapse a hand-rolled tool-calling loop into a few lines, and what that trade-off actually costs you.https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/04-graphs-not-chains/Mon, 10 Aug 2026 00:00:00 +0000https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/04-graphs-not-chains/Why LangGraph models an agent as an explicit graph of nodes and edges instead of a hidden loop, and the core primitives that make that possible.https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/05-durable-state-and-checkpoints/Mon, 10 Aug 2026 00:00:00 +0000https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/05-durable-state-and-checkpoints/How LangGraph snapshots a graph’s state after every step, so a run can survive a crash, get inspected mid-flight, or resume exactly where it left off.https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/06-human-in-the-loop-interrupts/Mon, 10 Aug 2026 00:00:00 +0000https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/06-human-in-the-loop-interrupts/How LangGraph’s interrupt() pauses a running graph mid-step to ask a human before a risky action, and how Command(resume=…) picks it back up.https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/07-structured-output-as-a-guardrail/Mon, 10 Aug 2026 00:00:00 +0000https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/07-structured-output-as-a-guardrail/How binding a Pydantic schema to a model or agent turns a probably-shaped response into a validated object, and what still isn’t guaranteed once it is.https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/08-a-minimal-eval-harness/Mon, 10 Aug 2026 00:00:00 +0000https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/08-a-minimal-eval-harness/A small, hand-rolled harness for checking whether an agent’s output is actually correct, not just well-formed, with a fixed dataset, a scorer per case, and a pass rate.https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/09-llm-as-judge/Mon, 10 Aug 2026 00:00:00 +0000https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/09-llm-as-judge/When correctness is subjective, grade agent output with a second, structured LLM call instead of eyeballing every run.https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/10-tracing-agents-with-opentelemetry/Mon, 10 Aug 2026 00:00:00 +0000https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/10-tracing-agents-with-opentelemetry/When one specific agent run goes wrong, aggregate pass rates won’t tell you where. A span tree of every model call and tool call will, whether it’s built automatically, by a decorator, or by hand.https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/11-timeouts-retries-and-loop-limits/Mon, 10 Aug 2026 00:00:00 +0000https://www.fernandohermida.com/posts/agentic-ai-engineering-journey/11-timeouts-retries-and-loop-limits/A timeout bounds a call, a retry survives a transient failure, and a hard iteration limit is what stops an agent that never planned to.