BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.devconf.info//devconf-cz-2026//talk//QUESU8
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-devconf-cz-2026-QUESU8@pretalx.devconf.info
DTSTART;TZID=CET:20260619T110000
DTEND;TZID=CET:20260619T113500
DESCRIPTION:We’ve all seen the demos: an AI builds a flashy app from scra
 tch\, and everyone asks\, "What prompt did you use?" But try that "magic p
 rompt" on a massive legacy system or deep inside Linux OS tools\, and it f
 alls apart. The "perfect prompt" is a myth. In real-world codebases\, agen
 ts suffer from "Context Amnesia" - forgetting architectural constraints as
  the chat grows. They also over-engineer\, duplicate code\, hallucinate an
 d add redundant fallbacks just to be safe.\n\nThis is why human engineers 
 are irreplaceable. We hold the big-picture context and the judgment of wha
 t not to build.\n\nThis talk cuts through the "autonomous AI" hype. Instea
 d of endless prompt-hacking\, I’ll share a practical\, design-first work
 flow for tools like Cursor and Claude Code. You will learn how to:\n\n- An
 chor Context: use Markdown design docs as the agent's "external memory."\n
 - Filter the Bloat: force agents to plan first\, catching garbage before i
 t's coded.\n- Steer\, Don't Prompt: recover when the agent gets stuck.
DTSTAMP:20260430T131532Z
LOCATION:E104 (capacity 72)
SUMMARY:Stop Looking for the Perfect Prompt: The Design-First Workflow for 
 Coding Agents - Mark Kemel
URL:https://pretalx.devconf.info/devconf-cz-2026/talk/QUESU8/
END:VEVENT
END:VCALENDAR