Wojtek Kuberski
where else to find them
linkedin:wojtek-kuberski
what they talk about
where they keep going back
- Open Source Summit Latin America2 editions2022–2022
all 4 in the filters
on a stage with
- Aditi Ramaswamyonce
- Andrew Knightonce
- Anisha Biswarayonce
all 13 in the filters
who talks about the same things
- Emeli Dral8 talks
- Tobias Sterbak4 talks
- Simona Maggio2 talks
- Jean-François Puget3 talks
- Gaurav Mittal3 talks
- Alejandro Saucedo33 talks
| title | conference | date |
|---|---|---|
| The ML Monitoring Flow for Models Deployed to Production | PyData Amsterdam 2024 | 2024-09-18 |
Models degrade in production, median ~20% error increase at 365 days and some >80%, so reliance on tests fails; univariate drift tests produce many false… | ||
| Hallucination-Free LLMs: Strategies for Monitoring and Mitigation | AI_dev: Open Source GenAI & ML Summit Europe 2024 | 2024-06-19 |
Describes practical monitoring and mitigation: two main families — LLM self-evaluation (use one LLM to score answers, e.g. SelfCheckGPT / LLM-eval) which needs… | ||
| How to Detect Silent Failures in ML Models | Open Source Summit Latin America 2022: Open AI + Data Forum | 2022-08-23 |
Open-source Python library NannyML implements a concrete method: it calibrates model probabilities, applies a decision threshold, builds per-row partial… | ||
| How to Detect Silent Failures in ML Models | Open Source Summit Latin America 2022 | 2022-08-23 |
Claims a practical way to estimate model performance without labels by calibrating predicted probabilities, turning each prediction into a partial confusion… | ||
| Premiere | Conf42 Machine Learning 2022 | 2022-05-19 |
Conference opens with the claim that machine learning won’t put anyone out of a job but will change how work gets done and then moves into concrete tools and… | ||
| How to detect silent failures in ML models | Conf42 Machine Learning 2022 | 2022-05-19 |
Defines two concrete failure modes: data drift (change in input distribution) and concept drift (change in true mapping), and shows concept drift almost always… | ||
| nothing matches | ||