BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:From Laptop to Supercomputer: Building Production ML Pipelines wit
 h dagster-slurm
DTSTART:20260417T090000Z
DTEND:20260417T100000Z
DTSTAMP:20260430T003000Z
UID:indico-event-293@events.asc.ac.at
CONTACT:training@ai-at.eu
DESCRIPTION:Date: 17 April 2026\, 11:00 – 12:00 (CEST) Price: Free (f
 or eligible participants)Format: AI:AT Webinar\, ONLINE (Zoom) Target au
 dience: Industry\, public admin\, academiaLanguage: English Organizer:
  AI:AT & ASCBridge the gap between data science prototypes and production
 -scale machine-learning (ML) workloads on HPC clusters. Learn how dagster-
 slurm brings modern orchestration to supercomputing environments without r
 ewriting your code and how this allows to use EU sovereign cost-efficient 
 GPUs.As machine learning (ML) models grow in complexity and datasets expan
 d beyond what a single machine can handle\, data scientists and ML enginee
 rs face a critical challenge: how to scale their work from a laptop to hig
 h-performance computing infrastructure without completely rewriting their 
 code or becoming HPC experts.This session introduces dagster-slurm\, an op
 en-source integration for running Dagster assets on Slurm HPC clusters. da
 gster-slurm lets you take the same Dagster pipelines from a laptop to a Sl
 urm-backed supercomputer with minimal configuration changes. As an additio
 nal benefit you can profit from EU sovereign cost efficient GPUs.We will e
 xplore:Why this matters: The gap between local development and HPC product
 ion\, and why traditional approaches create frictionHow it works: A live d
 emonstration showing the same pipeline running locally and on a real super
 computer (MUSICA with GPUs)Key capabilities: Environment reproducibility w
 ith pixi\, distributed computing with Ray\, monitoring\, and automatic dep
 endency managementThe session includes live demonstrations:Transforming a 
 simple Python script into a scalable HPC pipeline in minutesA production d
 ocument processing workflow with Docling + Ray on GPU infrastructure for R
 AG data preprocessingIdeally\, prior python experience is available. But w
 e'll introduce the concepts as we go. If you write Python and want to scal
 e your ML workloads beyond a single machine\, this session will show you a
  practical path forward.Target audienceEveryone is welcome who wants to sc
 ale data science and machine learning workflows beyond a single computer a
 nd benefit from EU sovereign compute.This webinar is particularly relevant
  for:Data Scientists & ML Engineers looking to leverage HPC resourcesMLOps
  Engineers seeking better orchestration for HPC workloadsResearch Software
  Engineers bridging academic and production systemsPlatform Engineers eval
 uating HPC orchestration solutionsThis webinar is open and free of charge 
 for all participants from academia\, industry\, and public administration 
 from EU and/or EuroHPC JU member countries.Entry level & prerequisitesInte
 rmediateBasic Python programming experience is required.Familiarity with M
 L/data processing workflows is helpful.No prior knowledge of HPC\, Slurm\,
  or Dagster is required.Agenda10:00   Welcome & Presentation of AI Factor
 y Austria AI:AT10:05   Introduction to dagster-slurm10:10   Live Demo 1:
  From Local Script to HPC Cluster10:20   Core Concepts: Environment Manag
 ement & Orchestration10:35   Live Demo 2: Production Pipeline with Doclin
 g + Ray on GPUs10:50   Questions & DiscussionSpeakerGeorg Heiler (ASCII S
 upply Networks & T-Mobile Austria)      Georg is a Senior Data Exper
 t at T-Mobile Austria (Magenta) and an Research Software Engineer at Compl
 exity Science Hub and ASCII Supply Networks\, where he solves complex chal
 lenges with data at scale.     At Magenta\, Georg leads the cloud mig
 ration of enterprise data platforms\, architecting modern data infrastruct
 ure for Austria's leading telecommunications provider. At ASCII\, he tackl
 es large-scale multi-modal ML-Ops challenges\, building systems that proce
 ss large datasets of supply chain data.     Georg is the creator and 
 lead maintainer of dagster-slurm\, an open-source framework connecting mod
 ern orchestration with high-performance computing. He also contributes to 
 metaxy\, a toolkit for supply chain knowledge graph construction. His work
  bridges the gap between cutting-edge research infrastructure and producti
 on-grade systems\, making supercomputing accessible to data science teams.
 Connect & Project Links:Personal Website: https://georgheiler.comdagster-s
 lurm GitHub: https://github.com/ascii-supply-networks/dagster-slurmmetaxy 
 GitHub: https://github.com/anam-org/metaxyDocumentation: https://dagster-s
 lurm.geoheil.comPresentation Slides: https://dagster-slurm.geoheil.com/sli
 des/                                     https://dagst
 er-slurm.geoheil.com/slides-multimodal/LanguageEnglishDate\, time\, and lo
 cation17.04.2026\, 11:00 – 12:00 (CEST)\, LIVE ONLINE WEBINAR (Zoom)Regi
 strationRegistration is required\, the link to the registration form can b
 e found on the top of this page in the menu on the left. Please register w
 ith your official institutional email address to prove your affiliation.Yo
 u will get the Zoom link in the automatic confirmation by email (subject s
 tarting with "[Indico] Registration")\, please check your Spam/Junk folder
 s). We will also send 1–2 reminders before the webinar.Please do not hes
 itate to contact us at training@ai-at.eu if you have any questions.Organiz
 ers                This webinar is jointly organized by AI Fac
 tory Austria AI:AT and Austrian Scientific Computing ASC (aka ASC Research
  Center\, TU Wien).AcknowledgementsAI Factory Austria AI:AT has received f
 unding from the European High-Performance Computing Joint Undertaking (JU)
  under grant agreement No 101253078. The JU receives support from the Hori
 zon Europe Programm of the European Union and Austria (BMIMI / FFG). \n\n
 https://events.asc.ac.at/event/293/
LOCATION:Zoom (ONLINE)
URL:https://events.asc.ac.at/event/293/
END:VEVENT
END:VCALENDAR