BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pydata-london-2026//talk//APYSNR
BEGIN:VTIMEZONE
TZID:GMT
BEGIN:STANDARD
DTSTART:20001029T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:GMT
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T020000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:BST
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-pydata-london-2026-APYSNR@pretalx.com
DTSTART;TZID=GMT:20260605T105000
DTEND;TZID=GMT:20260605T122000
DESCRIPTION:AI agents are moving into production in 2026\, but when somethi
 ng goes wrong (a tool call fails silently\, an LLM takes 13 seconds to res
 pond\, token costs spike overnight) teams struggle to diagnose issues acro
 ss multi-step agentic workflows. In this hands-on tutorial you will solve 
 a real problem on the island in Animal Crossing by building a Model Contex
 t Protocol (MCP) server in Python using FastMCP\, instrumenting it with Op
 enTelemetry following the emerging GenAI and MCP semantic conventions and 
 visualising end-to-end traces in a local Jaeger instance. Did I mention th
 at events on the island occur in real time and are collected and processed
  using Apache Kafka?\n\nYou will learn how distributed tracing captures th
 e hierarchical relationship between agent conversations\, tool executions 
 and MCP protocol messages\, and how to use that visibility for debugging\,
  cost analysis and performance optimisation (including picking the right m
 odel and checking if you’re drowning in serialisation overhead). You wil
 l leave with a fully instrumented MCP server\, a Docker Compose real-time 
 observability stack and the knowledge to bring production-grade observabil
 ity to your own agentic AI systems.
DTSTAMP:20260602T223155Z
LOCATION:Doddington Forum
SUMMARY:Observing Agentic AI in Production: MCP Server Tracing with OpenTel
 emetry and Animal Crossing - Tun Shwe\, Fei Phoon
URL:https://pretalx.com/pydata-london-2026/talk/APYSNR/
END:VEVENT
END:VCALENDAR