BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
DTSTART:20260308T030000
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:EDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20261101T010000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:EST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260914T215716Z
UID:7AA5CCE8-C014-4FB9-906F-9E43C5573890
DTSTART;TZID=America/New_York:20261013T180000
DTEND;TZID=America/New_York:20261013T200000
DESCRIPTION:[Subhash_140x211]Subhash Kovela\n\nAs AI-assisted software engi
 neering evolves beyond interactive chat and ad-hoc prompting into context 
 engineering\, Spec-Driven Development (SDD) has emerged as a crucial disci
 pline for generating predictable\, production-ready code. However\, moving
  from theory to implementation presents real challenges: Which spec framew
 orks actually hold up in practice? And how to adapt to a particular organi
 zation&#39;s coding best practices without overloading the context?\n\nIn this
  talk\, we examine three primary Opensource SDD frameworks: BMAD\, OpenSpe
 c\, and SpecKit. The speaker will walk through workflows to evaluate how e
 ach framework structures specs\, handles edge cases\, and manages executio
 n. Additionally\, we’ll analyze the trade-offs of using enriched context
  vs. lean context\, highlighting where additional context sharply reduces 
 hallucination and where it simply adds noise or degrades performance.\n\nT
 he goal is to help engineering teams make an informed choice by covering t
 hese areas:\n\n- Tool Trade-offs: Key usability differences between the th
 ree frameworks\n- Context Engineering: Practical rules for structuring con
 text to maximize generation quality without over-saturating the model.\n- 
 Gotchas: Where SDD tooling falls short and how to design resilient specs f
 or daily engineering workflows.\n\nCo-sponsored by: IEEE Boston AI Group\n
 \nSpeaker(s): Subhash Kovela\, \n\nVirtual: https://events.vtools.ieee.org
 /m/575332
LOCATION:Virtual: https://events.vtools.ieee.org/m/575332
ORGANIZER:info@voptronix.com
SEQUENCE:121
SUMMARY:Agentic coding and Spec-Driven Development in Practice
URL;VALUE=URI:https://events.vtools.ieee.org/m/575332
X-ALT-DESC:Description: &lt;br /&gt;&lt;table style=&quot;border-collapse: collapse\; wid
 th: 100%\; height: 280.4px\; border-width: 0px\;&quot; border=&quot;1&quot;&gt;&lt;colgroup&gt;&lt;co
 l style=&quot;width: 20.0713%\;&quot;&gt;&lt;col style=&quot;width: 79.9287%\;&quot;&gt;&lt;/colgroup&gt;\n&lt;t
 body&gt;\n&lt;tr style=&quot;height: 280.4px\;&quot;&gt;\n&lt;td style=&quot;height: 280.4px\; border
 -width: 0px\;&quot;&gt;\n&lt;p style=&quot;text-align: center\;&quot;&gt;&lt;img src=&quot;https://events.
 vtools.ieee.org/vtools_ui/media/display/755c8698-00ce-47ed-b1c4-345b0305aa
 a6&quot; alt=&quot;Subhash_140x211&quot; width=&quot;140&quot; height=&quot;211&quot;&gt;&lt;strong&gt;Subhash Kovela&lt;
 /strong&gt;&lt;/p&gt;\n&lt;/td&gt;\n&lt;td style=&quot;height: 280.4px\; border-width: 0px\;&quot;&gt;\n&lt;
 p&gt;As AI-assisted software engineering evolves beyond interactive chat and 
 ad-hoc prompting into context engineering\, &lt;strong&gt;Spec-Driven Developmen
 t (SDD)&lt;/strong&gt; has emerged as a crucial discipline for generating predic
 table\, production-ready code. However\, moving from theory to implementat
 ion presents real challenges: &lt;em&gt;Which spec frameworks actually hold up i
 n practice? And how to adapt to a particular organization&#39;s coding best pr
 actices without overloading the context?&lt;br&gt;&lt;/em&gt;&lt;/p&gt;\n&lt;p&gt;In this talk\, w
 e examine three primary Opensource SDD frameworks: &lt;strong&gt;BMAD&lt;/strong&gt;\,
  &lt;strong&gt;OpenSpec&lt;/strong&gt;\, and &lt;strong&gt;SpecKit&lt;/strong&gt;. The speaker wil
 l walk through workflows to evaluate how each framework structures specs\,
  handles edge cases\, and manages execution. Additionally\, we&amp;rsquo\;ll a
 nalyze the trade-offs of using enriched context vs. lean context\, highlig
 hting where additional context sharply reduces hallucination and where it 
 simply adds noise or degrades performance.&lt;/p&gt;\n&lt;/td&gt;\n&lt;/tr&gt;\n&lt;/tbody&gt;\n&lt;/
 table&gt;\n&lt;p&gt;The goal is to help engineering teams make an informed choice b
 y covering these areas:&amp;nbsp\;&lt;/p&gt;\n&lt;ul type=&quot;disc&quot;&gt;\n&lt;li style=&quot;mso-list:
  l0 level1 lfo1\; tab-stops: list .5in\;&quot;&gt;&lt;strong&gt;Tool Trade-offs:&lt;/strong
 &gt; Key usability differences between the three frameworks&lt;/li&gt;\n&lt;li style=&quot;
 mso-list: l0 level1 lfo1\; tab-stops: list .5in\;&quot;&gt;&lt;strong&gt;Context Enginee
 ring:&lt;/strong&gt; Practical rules for structuring context to maximize generat
 ion quality without over-saturating the model.&lt;/li&gt;\n&lt;li style=&quot;mso-list: 
 l0 level1 lfo1\; tab-stops: list .5in\;&quot;&gt;&lt;strong&gt;Gotchas:&lt;/strong&gt; Where S
 DD tooling falls short and how to design resilient specs for daily enginee
 ring workflows.&lt;/li&gt;\n&lt;/ul&gt;
END:VEVENT
END:VCALENDAR