Luca Baggi
where else to find them
what they talk about
all 5 in the filters
where they keep going back
- PyCon Italia3 editions2023–2025
- PyData London3 editions2024–2026
- PyData Global2 editions2023–2024
all 4 in the filters
on a stage with
never shared a talk
who talks about the same things
- Matt Harrison16 talks
- Vincent D Warmerdam22 talks
- Erik Erlandson14 talks
- Jodie Burchell18 talks
- (unnamed presenter)4 talks
- Ian Ozsvald17 talks
| title | conference | date |
|---|---|---|
| Reading the Mind of an LLM | PyData London 2026 | 2026-06-05 |
Claims that mechanistic interpretability can locate and manipulate feature directions inside real LLMs — using sparse dictionary learning, circuit tracing and… | ||
| LLM Inference Arithmetics | PyData London 2025 | 2025-06-06 |
Explains how parameter counts map to VRAM and compute so numbers mean something: Gemma 2B is closer to a 3B as a parameter-vs-storage example, a 7B model… | ||
| Foundational Models for Time Series Forecasting: are we there yet? | PyCon Italia 2025 | 2025-05-28 |
Explains that transformer-based foundational forecasters are promising but data-limited: teams adapt T5-style and decoder transformers, bin or patch continuous… | ||
| LLM Inference Arithmetics: the Theory behind Model Serving | PyCon DE & PyData 2025 | 2025-04-23 |
Argues that transformer inference is inherently serial while training is parallel, so training needs ~3× the FLOPs of inference and attention compute scales… | ||
| Foundational Models for Time Series Forecasting: are we there yet? | PyData Global 2024 | 2024-12-03 |
Transformers get repurposed for forecasting by quantize to discrete tokens (Chronos uses GPT‑2/T5 with mean scaling and binning) or by patching continuous… | ||
| Uncertainty estimation at scale with functime | PyData London 2024 | 2024-06-14 |
functime claims to produce prediction intervals for thousands of series locally by combining Polars-powered feature extraction and cross‑validation that runs… | ||
| Embeddings, Transformers, RLHF: Three key ideas to understand ChatGPT | PyCon Italia 2024 | 2024-05-22 |
Explains that large language models first learn dense embeddings (word vectors from Word2Vec-style masked learning) that let arithmetic-like analogies emerge… | ||
| functime: a next generation ML forecasting library powered by Polars | PyCon Italia 2024 | 2024-05-22 |
functime is a Python library that trains a single estimator across a panel to forecast thousands of time series on a laptop by using Polars (Rust + Rayon) for… | ||
| How I used Polars to build built functime, a next gen ML forecasting library | PyData Global 2023 | 2023-12-06 |
Claims a concrete speedup: feature engineering runs 1-2 orders of magnitude faster using Polars; batch inference and scaling happen without PySpark. Presents… | ||
| Rust is easy | PyCon Italia 2023 | 2023-05-25 |
Claim: Rust is easy if you trust the compiler. The demo installs and uses rustup and cargo, edits in Vim with rust-analyzer and Copilot disabled, drops a… | ||
| Is the great dataframe showdown finally over? Enter: polars | PyCon Italia 2023 | 2023-05-25 |
Polars claims multi threaded, Arrow based execution with a query planner that does predicate pushdown and lazy scan to lazy frame streaming so users can… | ||
| nothing matches | ||