BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250822T115805Z
LOCATION:Room 5.2D02
DTSTART;TZID=Europe/Stockholm:20250617T150000
DTEND;TZID=Europe/Stockholm:20250617T153000
UID:submissions.pasc-conference.org_PASC25_sess129_msa242@linklings.com
SUMMARY:Leveraging AI-Driven Code Generation for Portable and Scalable Sim
 ulations
DESCRIPTION:Osman Seckin Simsek and Florina Ciorba (University of Basel)\n
 \nAs high performance computing (HPC) applications continue to push the bo
 undaries of complexity and performance, leveraging AI-driven code generati
 on has emerged as a powerful approach to streamline scientific software de
 velopment processes.\nThis talk explores the role of AI code generation mo
 dels as a collaborative pair programmer in developing portable and scalabl
 e code with the SPH-EXA simulation framework.\nSpecifically, we use AI-bas
 ed code generation to assist development of SPH-EXA using CUDA, HIP, and S
 YCL to automate the code implementation and unit test generation and signi
 ficantly reduce development effort while maintaining high standards of cor
 rectness and performance. \nOur work evaluates the dynamic interplay betwe
 en human expertise and generative AI capabilities. We will assess producti
 vity gains and discuss the practical challenges of ensuring performance po
 rtability and scalability across diverse hardware platforms. Through detai
 led performance analyses we will highlight both the benefits and the limit
 ations of current AI-assisted approaches for scientific software developme
 nt.\nThis talk will provide insights into how AI-driven tools can serve as
  effective collaborative partners in HPC code development.\n\nDomain: Comp
 utational Methods and Applied Mathematics\n\nSession Chair: Florina Ciorba
  (University of Basel)\n\n
END:VEVENT
END:VCALENDAR
