Send us a text

Ever wondered how cycling from London to Paris could inspire an in-depth discussion aboutAI? Join us as Marc shares his scenic adventure along the charming Avenue Vert, setting the stage for a captivating exploration into the world of open-source large language models (LLMs). Christian and Marc delve into the heart of AI, examining the strengths and limitations of open models like Mistral, Meta's Llama, and Google's Gemma compared to closed systems such as GPT, Gemini or Claude. Discover the power of open weights and how they provide users with the flexibility to harness pre-trained parameters for a myriad of applications, shedding light on the architecture and performance of these rapidly evolving models.

As we navigate the nuanced landscape of AI, we compare the practicalities and strategic advantages of open versus closed LLMs. While open-source models may require a steeper learning curve and higher initial investment in technical expertise, they offer unparalleled customisation for specialised tasks and niche programming languages. Conversely, closed models like GPT-4 and Gemini shine in delivering cost-effective solutions for more general applications. We also delve into the strategic deployment of LLMs for on-device applications and the critical importance of data sovereignty in fields like healthcare. Tune in to gain a comprehensive understanding of how these developments might impact your projects and reshape your industry.

Follow us on LinkedIn Instagram Twitter TikTok and Youtube !
Music by Skilsel.
This podcast represents our views and not the ones of our employers.

Grab your mug, it's Data Coffee Break time with Christian and Marc! ☕️

This episode, we're brewing up a hot topic: RAG vs. Giant Context Windows. You might be wondering, what in the world is RAG and why should I care? Well, settle in because we're about to break it down for you, no matter if you're an AI rockstar or just dipping your toes into the data pool.

We'll be chatting about the latest news in the wild world of AI, then diving deep into this whole RAG situation. Is the future all about these Large Context Window models with millions of tokens, or will RAG still be the king of the castle? We'll explore the pros and cons of each approach, so you can see which one might be the better fit for your next project.

Plus, we won't just leave you hanging with theory! We'll also be uncovering the real-world uses for both RAG and LCLs. Think of it like peeking into the future and seeing how these technologies will change the game!

So, grab your favourite drink, hit play, and get ready for a stimulating conversation that will leave you hyped about the future of AI,

Follow us on LinkedIn Instagram Twitter TikTok and Youtube !
Music by Skilsel.
This podcast represents our views and not the ones of our employers.

Send us a text

One year after our first attempt to predict the data trends of 2023, we look back into our prediction and reflect on how this year was transformative.
Then, we gear back into 2024 with a heated debate with Ivan Coelho, our infamous guest, to dissect what could be a transformative year for the Data and AI space.
Tune in and listen to our take for 2024 and some tips along the way.

Some interesting reading:
https://sloanreview.mit.edu/article/five-key-trends-in-ai-and-data-science-for-2024/

https://towardsdatascience.com/trends-that-will-shape-the-modern-data-stack-in-2024-6b7de28335c2

https://www.datacamp.com/blog/the-top-5-vector-databases

https://towardsdatascience.com/navigating-the-ai-landscape-of-2024-trends-predictions-and-possibilities-41e0ac83d68f#b80d

https://blog.eliassen.com/projecting-the-top-trends-in-data-in-motion

Follow us on LinkedIn Instagram Twitter TikTok and Youtube !
Music by Skilsel.
This podcast represents our views and not the ones of our employers.