BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//aiweek.boston//Schedule//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Building an AI Software Factory with a Human in the Loop
X-WR-TIMEZONE:America/New_York
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20070311T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20071104T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:building-an-ai-software-factory-with-a-human-in-the-loop@aiweek.boston
DTSTAMP:20261003T062352Z
SEQUENCE:0
STATUS:CONFIRMED
DTSTART;TZID=America/New_York:20261015T130000
DTEND;TZID=America/New_York:20261015T134500
SUMMARY:Building an AI Software Factory with a Human in the Loop
DESCRIPTION:In 1801 Eli Whitney assembled muskets from a pile of intercha
 ngeable parts in front of Congress\, and became famous for it. The demons
 tration was staged\; the parts were hand fitted and secretly marked. \n\n
 Real interchangeability took another forty years... if that sounds like a
  few AI demos you have seen lately\, this talk is for you.\n\nSoftware fa
 ctories are not a 2024 idea. People have been trying to build them since 
 1968\, and why those attempts failed is more useful than why we think thi
 s one will work. Sid Probstein\, founder of conjure.work\, has been runni
 ng an agentic software factory in production: fleets of AI agents that de
 sign\, build and test real software\, directed by senior engineers\, with
  a person accountable at three fixed points.\n\nThis is a working session
  on what actually holds it together\, and it is not the agents.\n\nWe wil
 l cover:\n\n* Why the old discipline still works and is harder to apply. 
 Work breakdown structures\, RACI\, retrospectives: none of it stopped bei
 ng true because the workers became agents.\n\n* Why narrowing what an age
 nt is responsible for buys more reliability than making it smarter.\n\n* 
 Why speed without proof is not leverage\, and what that demands of CI/CD.
 \n\n* Where the human has to stand. Toyota gave every worker on the line 
 a cord that stops it\, and the line got faster. That remains the best ans
 wer anyone has found.\n\n* What a feedback loop looks like when your user
 s' own words decide what gets built next.\n\n* Where this goes next\, inc
 luding the research question nobody has answered: how little human oversi
 ght is enough\, and where in the pipeline it has to sit.\n\n45 minutes: 3
 0 minutes of talk\, 15 of questions. Free. Virtual.
LOCATION:https://luma.com/5jkb17ry
URL:https://aiweek.boston/schedule/building-an-ai-software-factory-with-a
 -human-in-the-loop
BEGIN:VALARM
ACTION:DISPLAY
DESCRIPTION:Building an AI Software Factory with a Human in the Loop
TRIGGER:-P1D
END:VALARM
BEGIN:VALARM
ACTION:DISPLAY
DESCRIPTION:Building an AI Software Factory with a Human in the Loop
TRIGGER:-PT1H
END:VALARM
END:VEVENT
END:VCALENDAR