William Benton
where else to find them
no profile links on file
what they talk about
all 8 in the filters
where they keep going back
- Berlin Buzzwords8 editions2018–2026
- DevConf.CZ2 editions2019–2020
on a stage with
all 4 in the filters
who talks about the same things
- Sophie Watson10 talks
- Tom Coufal7 talks
- Salman Iqbal11 talks
- Clayton Coleman12 talks
- Cedric Clyburn22 talks
- Ivan Pedrazas5 talks
| title | conference | date |
|---|---|---|
| Sunset for the Wild West: Making ML disciplined by default | Berlin Buzzwords 2026 | 2026-06-07 |
Claims that dependency chaos makes ML systems insecure and unreproducible, supported by a PyTorch container dependency graph with ~1,500 nodes and ~6,000… | ||
| "Do What I Mean": The History of AI and Program Synthesis | Berlin Buzzwords 2025 | 2025-06-15 |
A concise lineage from 50 years of program synthesis to today's LLM coding shows concrete tools and results: FlashFill and successor FlashMeta use backward… | ||
| Large language models are not a paradigm shift | Berlin Buzzwords 2024 | 2024-06-09 |
Argues that LLMs are not a paradigm shift for computing but a new, powerful instance of familiar ML systems, showing concrete analogies: fourth‑order Markov… | ||
| Synthetic data: when, why, and how | Berlin Buzzwords 2023 | 2023-06-18 |
Argues synthetic tabular data is practical for privacy, load testing, and rare-event simulation, showing concrete techniques: sampling marginals, sampling… | ||
| Luxuries, necessities, and the challenges that remain | Berlin Buzzwords 2022 | 2022-06-12 |
Shows concrete results using Nvidia hardware and Python: cuPy and RAPIDS let UMAP run at interactive latency so visual exploration uses full data instead of… | ||
| Facets of Reproducible research on Kubernetes | Berlin Buzzwords 2021 | 2021-06-15 |
Kubernetes plus CI/CD and container images make environments reproducible, but real reproducibility requires concrete steps: use binder or declarative images… | ||
| How to avoid columnar calamities: what no one told you about Apache Parquet | Berlin Buzzwords 2021 | 2021-06-15 |
Parquet wins size and speed by columnar layout, run-length and dictionary encodings, and predicate pushdown; a cited example compresses 50M synthetic payments… | ||
| Cloud Native Machine Learning Systems at Day Two and Beyond | KubeCon + CloudNativeCon North America 2020 Virtual | 2020-11-17 |
Kubernetes becomes the default infra for ML but it leaves gaps around reproducibility, interactive velocity, and model monitoring; expect to use CI/CD (Tekton)… | ||
| Keynote: Solving the problems of AI in production | DevConf.CZ 2020 | 2020-01-24 |
Argues that model training is easy but production ML fails on integration, not algorithms: cites adversarial fragility of deep nets and a 250,000 GPU‑hour… | ||
| Band-Aids don’t fix bullet holes | Berlin Buzzwords 2019 | 2019-06-16 |
Argues that production ML pain comes from pipeline entanglement and vendor-driven point solutions, not lack of models: deep learning excels at vision/speech… | ||
| Radanalytics.io streaming applications workshop | DevConf.CZ 2019 | 2019-01-25 |
Shows a concrete pipeline: synthetic social updates (Jane Austen + reviews) stream into Kafka, Apache Spark Structured Streaming applies a spaCy + VADER… | ||
| Data sketching workshop | DevConf.CZ 2019 | 2019-01-25 |
Workshop demonstrates practical sketches: Bloom filter for approximate set membership with mergeable bitmaps and a false‑positive curve that climbs toward 1 as… | ||
| Container workflows for data science and ML | DevConf.CZ 2019 | 2019-01-25 |
Claims OpenShift is a powerhouse for ML and shows concrete tooling and demos: uses OpenShift and radanalytics.io to spin up an app-scoped Spark cluster, uses… | ||
| Intelligent applications in containers | Berlin Buzzwords 2018 | 2018-06-11 |
Shows a concrete workflow: prototypes run interactively as Jupyter + Spark notebooks (k‑means, logistic regression, Word2Vec) then get lifted into a… | ||
| nothing matches | ||