Luca Baggi

As Large Language Models (LLMs) transition into core enterprise infrastructure, treating them as "black boxes" is no longer an acceptable engineering risk. Can we turn LLM agents into debuggable, predictable software systems?

What if you could watch an AI’s thought take shape? For years, LLMs have been impenetrable "black boxes," but we are finally beginning to find ways to see how the ghost in the machine actually works.

What if you could watch an AI’s thought take shape? For years, LLMs have been impenetrable "black boxes," but we are finally beginning to find ways to see how the ghost in the machine actually works.

With o1, OpenAI ushered a new era: LLMs with reasoning capabilities. This new breed of models broadened the concept of scaling laws, shifting focus from train-time to inference-time compute. But how do these models work? What do we think their architectures look like, and what data do we use to...

What if you could watch an AI’s thought take shape? For years, LLMs have been impenetrable "black boxes," but we are finally beginning to find ways to see how the ghost in the machine actually works. This talk explores **mechanistic interpretability**, a subfield of AI that aims to understand...

Have you ever asked yourself how parameters for an LLM are counted, or wondered why Gemma 2B is actually closer to a 3B model? You have no clue about what a KV-Cache is?

Transformers have changed fields from Computer Vision to protein-folding. But what about forecasting? Come learn how titans like Amazon and Google train their foundational models, and, most importantly, how not to fall for the hype and learn how to read the public benchmarks.

Have you ever asked yourself how parameters for an LLM are counted, or wondered why Gemma 2B is actually closer to a 3B model? You have no clue about what a KV-Cache is?

functime is a modern time-series forecasting library to generate predictions for thousands of time series at once, while never leaving your laptop. Thanks to Polars' powerful query engine, feature extraction and cross-validation are 1-2 orders of magnitude faster.

Everyone is using ChatGPT, but few know how it works. This talk will drive the audience through the main ideas powering ChatGPT, to rip through the curtain of magic around the tool.

Powered by this impressive query engine, functime enables **forecasting thousands of time series at once, from the comfort of your laptop**. With Polars, we can push the boundary for what "reasonable scale" means - and build a **new generation of tools for machine learning**.

Polars conquered dataframes, and now it is coming for machine learning! With Polars-powered feature-extraction and a best-of-the-class set of diagnostic tools, functime enables **forecasting thousands of time series all at once, from the comfort of your laptop**. Though forecasting practitioners...

Every dataframe library that came after `pandas` promised an expressive API _and_ better performance, yet none replaced the fluffy bamboo-eater 🐼 Or has it? Enter `polars`: will its multi-threaded, in-memory query engine be enough to dethrone the king? Come and learn to tame this new artic beast!