PLN

PLN

Bonjour !

I'm PLN, and I love to transform…

…code into solutions to human problems


At Algolia, I've crafted tools that help you build mobile apps, understand your users, and show them your world.
Nowadays, I create GenAI tools for builders of generative experiences
blending search engines and language models together.

…code into music

As ParVagues, I write patterns that shape soundwaves.
I share those as tracks, videos, and live performances.

…code into animated pixelscapes

Using Hydra, I create animations that offer windows into other worlds.

…words into poems

A selection of texts scattered across note books and note apps.

…ideas into talks

I speak about topics I care about: from sharing my passions and teaching useful patterns, to questioning our current mental models.


Most recent talks
  • Search Relevance in the Gen AI Age You’re just done tuning search relevance perfectly for your human users, and just when you think you’re done, now you must do it again for robots? This is the age of RAG: we return relevant search results, not to a user, but to a LLM. What does this change? Can I reuse what I already know about relevance? You’ll learn that relevance for LLMs is not far from relevance for human users, and most of your hard work can be reused!
  • La French Stack: Search Relevance et GenAITu viens de peaufiner la pertinence de recherche pour tes utilisateurs humains, et juste quand tu crois en avoir fini, voilà qu'il faut recommencer pour les robots ? Bienvenue à l'ère du RAG : on renvoie des résultats pertinents, non plus à un humain, mais à un LLM. Qu'est-ce que ça change ? Peut-on réutiliser ce qu'on sait déjà sur la pertinence d'un résultat de recherche ? On va voir que la pertinence pour les LLMs n'est pas si différente des critères humains, et que la plupart de ton travail acharné peut être réutilisé !
  • Panel on RAG (Retrieval Augmented Generation)Roundtable with Pierre Peigné, Chief Science Officer at PRISM Eval & Alessandro Benedetti, Director at Sease, on Retrieval-Augmented Generation: why it's relevant today, what are the potential benefits and risks, and how to do it properly to capture most of its value in your projects.