<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>OpenTelemetry on Anton Chirikalov's Blog</title><link>https://antonch.me/en/tags/opentelemetry/</link><description>Recent content in OpenTelemetry on Anton Chirikalov's Blog</description><generator>Hugo -- 0.164.0</generator><language>en-US</language><lastBuildDate>Thu, 06 Aug 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://antonch.me/en/tags/opentelemetry/index.xml" rel="self" type="application/rss+xml"/><item><title>Agentic AI: the agent engineer's stack</title><link>https://antonch.me/en/posts/agentic-ai-stack/</link><pubDate>Thu, 06 Aug 2026 00:00:00 +0000</pubDate><guid>https://antonch.me/en/posts/agentic-ai-stack/</guid><description>A walk through the agentic development stack: what an agent actually is and how it differs from a chain of calls, PydanticAI in detail, LangGraph, smolagents and CrewAI, the MCP and A2A protocols, agent evaluation with output and trajectory metrics, the LLM judge and its biases, DeepEval with code and scenarios, an evaluation pipeline from scratch, RAG and HNSW, observability with OpenTelemetry</description></item></channel></rss>