Code for Thought

Peter Schmidt

Welcome to Code for Thought, the podcast about software for research and the people who make it.  Languages: English, German, French

  1. 5 Nov

    [EN] ByteSized: Green(er) Continuous Integration/Deployment - Joe Wallwork

    English Edition: In this edition of the ByteSized dRTP (STEP-UP) season Joe Wallwork from the Institute of Computing for Climate Science (https://iccs.cam.ac.uk/) is telling us why green computing matters. And in particular, how to make continuous integration/deployment more sustainable.  Links: https://github.com/Cambridge-ICCS/green-ci  The Green CI/CD GitHub toolhttps://cambridge-iccs.github.io/green-ci/  the documentation about ithttps://www.ebi.ac.uk/training/events/byte-sized-green-computing/ The whole set of courses on green computing by the EBML-EBI and Green Special Interest Group (see link below)https://socrse.github.io/green-sig/Some literature: https://arxiv.org/abs/2310.18718 Carbon-Awareness in CI/CDhttps://lorojournals.com/index.php/emsj/en/article/view/1581 Carbon-Aware Continuous Integration: Reducing Emissions Without Sacrificing Performancehttps://zenodo.org/records/12754189 Environmentally-aware use of GitHub Actionshttps://arxiv.org/abs/2510.26413 Environmental Impact of CI/CD PipelinesTools: https://codecarbon.io Code Carbonhttps://www.green-algorithms.org Green Algorithmshttps://github.com/GreenScheduler/cats CATS schedulerhttps://www.nextflow.io Nextflowhttps://app.electricitymaps.com/map/fifteen_minutes Electricity Map appGet in touch Thank you for listening! Merci de votre écoute! Vielen Dank für´s Zuhören! Contact Details/ Coordonnées / Kontakt: Email mailto:peter@code4thought.orgUK RSE Slack (ukrse.slack.com): @code4thought or @piddie Bluesky: https://bsky.app/profile/code4thought.bsky.socialLinkedIn: https://www.linkedin.com/in/pweschmidt/  (personal Profile)LinkedIn: https://www.linkedin.com/company/codeforthought/ (Code for Thought Profile)This podcast is licensed under the Creative Commons Licence:  https://creativecommons.org/licenses/by-sa/4.0/

  2. 14 Oct

    [EN] ByteSized: AI, agents and coding

    English edition: As part of our ByteSized series, we are going through some of the (many) aspects of AI assisted coding with my guests Robert Chatley (Imperial College London, UK), Reece Walsh (Newcastle Uni., UK) and Faruk Diblen (eScience Centre, Amsterdam, NL).  Links: https://missing.csail.mit.edu/2026/agentic-coding/ A good run-through of agentic coding by MIT/USAhttps://martinfowler.com/bliki/AgenticProgramming.html Martin Fowler on agentic codinghttps://github.blog/ai-and-ml/generative-ai/agent-pull-requests-are-everywhere-heres-how-to-review-them/ here's GitHub's suggestions on how to deal with AI generated pull requestshttps://github.com/fdiblen/rseng-agent-skills Faruk Diblen's set of SKILLS for research software engineeringhttps://ai-declaration.org and here is Faruk's AI declaration website and toolhttps://modelcontextprotocol.io/docs/2026-07-28/getting-started/intro Model Context Protocol (MCP) by Anthropichttps://rdr.ucl.ac.uk/articles/report/AI-Assisted_Software_Development_Policy_for_UCL_Advanced_Research_Computing/34070631?file=69660984 an example policy of how to deal with AI assisted coding tools - from UCL (University College London, UK)ByteSized dRTP is part of the STEP-UP project and I would like to thank them for supporting this show. Get in touch Thank you for listening! Merci de votre écoute! Vielen Dank für´s Zuhören! Contact Details/ Coordonnées / Kontakt: Email mailto:peter@code4thought.orgUK RSE Slack (ukrse.slack.com): @code4thought or @piddie Bluesky: https://bsky.app/profile/code4thought.bsky.socialLinkedIn: https://www.linkedin.com/in/pweschmidt/  (personal Profile)LinkedIn: https://www.linkedin.com/company/codeforthought/ (Code for Thought Profile)This podcast is licensed under the Creative Commons Licence:  https://creativecommons.org/licenses/by-sa/4.0/

  3. 12 Oct

    [EN] From research to startup: scikit-learn and :probabl.

    English Edition: scikit-learn is THE open source programming library for all things machine-learning. And since 2024, it is the core part of a company called :probabl.  What this means for the millions of users of scikit-learn and the many contributors is something we find out in this episode, where I talk to Marie Sacksick (skore), Arturo Amor (skolar) and Yann Lechelle (president and co-founder of :probabl.).  Links: https://www.probabl.ai Home page for :probabl. https://probabl.ai/skore  Skore https://github.com/probabl-ai/skore GitHub repo for skorehttps://scikit-learn.org/stable/about.html scikit-learn Python libraryhttps://blog.probabl.ai/intel-and-probabl-announce-collaboration-to-accelerate-scikit-learn-on-intel-hardware Collaboration between Intel and :probabl. https://scikit-learn-central.probabl.ai/#/catalog The central cataloguehttps://github.com/probabl-ai/scikit-learn-central GitHub repohttps://skolar.probabl.ai/ skolar, the certification programmehttps://inria.github.io/scikit-learn-mooc/ the INRIA MOOC for scikit-learnhttps://skrub-data.org/stable/ skrub (another project at :probabl.)https://github.com/skrub-data/skrub    Get in touch Thank you for listening! Merci de votre écoute! Vielen Dank für´s Zuhören! Contact Details/ Coordonnées / Kontakt: Email mailto:peter@code4thought.orgUK RSE Slack (ukrse.slack.com): @code4thought or @piddie Bluesky: https://bsky.app/profile/code4thought.bsky.socialLinkedIn: https://www.linkedin.com/in/pweschmidt/  (personal Profile)LinkedIn: https://www.linkedin.com/company/codeforthought/ (Code for Thought Profile)This podcast is licensed under the Creative Commons Licence:  https://creativecommons.org/licenses/by-sa/4.0/

Ratings & Reviews

5
out of 5
2 Ratings

About

Welcome to Code for Thought, the podcast about software for research and the people who make it.  Languages: English, German, French