IT Lightning Talks: session #26

Friday 9 May 2025, 10:00 → 11:08 Europe/Zurich

31/3-004 - IT Amphitheatre (CERN)

Pedro Ferreira (CERN), Sebastian Lopienski (CERN), Hugo Gonzalez Labrador (CERN), Elizaveta Ragozina (Massachusetts Inst. of Technology (US))

IT Lightning Talks (ITLT) are short presentations on any topic related to computing technology or to the IT department. See more here: https://twiki.cern.ch/IT/LightningTalks/

10:00 → 10:03 Welcome

Speakers: Elizaveta Ragozina (Massachusetts Inst. of Technology (US)), Hugo Gonzalez Labrador (CERN), Pedro Ferreira (CERN), Sebastian Lopienski (CERN)

2025.05_ITLT_session_26.pdf

2025.05_ITLT_session_26.pptx

Recording

Video preview

10:03 → 10:11 OpenMediaVault Storage Solution

I am currently exploring the OpenMediaVault NAS (Network Attached Storage) solution, and I have set up a personal mini-cloud solution for family and friends into a Raspberry Pi.

Speaker: Alexandros Petridis

IT_Lightning_NAS.pptx

Recording

Video preview

10:11 → 10:19 SolidJS and React.js: Two Paths to Reactive UI

At first glance, SolidJS and React.js code appears strikingly similar. However, their underlying mechanisms are vastly different. React.js relies on the Virtual DOM, while SolidJS employs a direct, fine-grained reactivity model, resulting in better performance. In this brief presentation, I'll highlight these key differences and similarities, aiming to introduce React enthusiasts to the performance benefits of SolidJS and spark interest in the broader concept of the reactivity.

Speaker: Karol Jozef Kosek

Google Slides Presentation

Google Slides Presentation

karol-kosek-en-im-amm-solid-react-presentation.pdf

karol-kosek-en-im-amm-solid-react-presentation.pptx

Recording

Video preview

10:19 → 10:27 Running Local LLMs with Ollama, Open-WebUI, and Browser-Use

Running large language models (LLMs) on your own machine is more accessible than you might think. In this talk, I’ll explain how to use Ollama to run LLMs locally, Open-WebUI as a user-friendly interface for interacting with them, and Browser-Use for working with agent-based LLMs. Installation, model selection, and practical applications like chat interfaces, automation, and local AI-assisted tasks will be covered

Speaker: Dr Igor Jakovljevic (CERN)

Recording

RUnning Local LLMs with Ollama, OpenWebUi, and Browser-Use.pdf

RUnning Local LLMs with Ollama, OpenWebUi, and Browser-Use.pptx

Video preview

10:27 → 10:35 From Netflix to Native: Automating French with Sentence Mining

What if binge-watching could make you bilingual? In this lightning talk, I’ll share how I automated my journey to learning French using a technique called sentence mining. By combining tools like subtitle extraction, spaced repetition, and smart automation, I turned every video into a personalized language lesson — all without ever opening a textbook. Whether you're learning a language yourself or curious about building educational workflows, this talk will show you how tech can turn passive watching into active learning.

Speaker: Alexandra Murariu

franki.pdf

Recording

Video preview

10:35 → 10:43 Double your work done: Typing fast & using shortcuts

How much do you think you type each day — 500 words, 1,000, 10,000? For many of us, using a keyboard has become one of the most essential parts of our daily work. And what’s better than doing something well? Doing it well and fast!

In this talk, I’ll share practical ways to boost your efficiency when using a computer — both by increasing your raw typing speed and by mastering keyboard shortcuts, which are especially valuable for repetitive tasks.

While the shortcut tips will focus on the Windows operating system, many also apply to macOS, Linux-based systems, and especially across web browsers. Tune in for the chance to learn something that you may actually use every single day!

Speaker: Siavas Firoozbakht (CERN)

Double your work done - Typing fast and using shortcuts.pdf

Double your work done - Typing fast and using shortcuts.pptx

Recording

Video preview

10:43 → 10:51 Mosh: Surviving >70% packet loss in the CLI

Have you ever roamed from building 31 to 513 and lost all your SSH sessions? Ever had to debug a problem remotely with a spotty mobile connection? Do you live in St Genis? In this talk I will discuss some of the common pitfalls that Modern SSH can encounter, while also showing Mosh, an alternate remote access protocol + shell that attempts to alleviate some of these problems.

Speaker: Zachary Goggin

make-slow.sh

mosh-talk.odp

mosh-talk.pdf

Recording

Video preview

10:51 → 10:59 Permacomputing - An approach to computing for hundreds of years

Modern computing suffers from a severe short sightedness. Most developments are driven by quarterly results, our habits become ever more computationally expensive.

In an age of environmental crisis, is there a different way to think of how we compute?

This talk gives an introduction to Permacomputing - a young philosophy and toolbox for computing over the span of hundreds of years.

Speaker: Richard Bachmann (CERN)

Permacomputing.pdf

Recording

Video preview