Michelle Frost
where else to find them
bluesky:aiwithmichelle.com
linkedin:michelle-frost-dev
what they talk about
all 5 in the filters
where they keep going back
- NDC London3 editions2024–2026
all 4 in the filters
on a stage with
never shared a talk
who talks about the same things
- Kesha Williams7 talks
- Vincent D Warmerdam22 talks
- Katharine Jarmul25 talks
- Jigyasa Grover8 talks
- Fiona Coath4 talks
- Tess Ferrandez-Norlander24 talks
| title | conference | date |
|---|---|---|
| Between the Layers– Interpreting Large Language Models | NDC London 2026 | 2026-01-26 |
Survey of recent interpretability work argues that mechanistic features replace single-neuron stories: Anthropic’s 2023–2024 papers use sparse autoencoders to… | ||
| Between the Layers– Interpreting Large Language Models | NDC AI 2025 | 2025-11-11 |
Distinguishes explainability (outside, human accounts) from interpretability (inside, causal wiring) and reports concrete methods: unmixing activations with… | ||
| AI vs Humans: Rewriting the Rules of Success | NDC London 2025 | 2025-01-27 |
Tracing AI milestones and measurement, the talk argues current success metrics prioritize systems that beat humans (Deep Blue 1997, AlphaGo 2016, ChatGPT 2022)… | ||
| AI vs Humans Rewriting the Rules of Success | Build Stuff 2024 | 2024-11-13 |
History-first survey traces myths to modern systems and argues current success metrics fail, proposing multi‑axis evaluation instead of a single intelligence… | ||
| The Elephant in your Dataset Addressing Bias in Machine Learning | Build Stuff 2024 | 2024-11-13 |
Overview of measurable failures and fixes: explains that the Compass recidivism dataset yields similar overall accuracy but Black defendants receive false… | ||
| The Elephant in your Dataset: Addressing Bias in Machine Learning | CPH DevFest 2024 | 2024-08-26 |
Walks through the ProPublica COMPAS dataset (~7,214 rows), reproduces the recidivism score with a simple logistic regression (~76% accuracy) and shows Black… | ||
| The Elephant in your Dataset: Addressing Bias in Machine Learning | NDC London 2024 | 2024-01-29 |
Defines seven injection points for harmful bias (historical, representation, measurement, aggregation, learning, evaluation, deployment) citing a 2021 paper… | ||
| nothing matches | ||