Conf42 Machine Learning 2022
2022-05-19
-- online --
[Conf42 Machine Learning 2021, Conf42 Machine Learning 2022, Conf42 Machine Learning 2023, Conf42 Machine Learning 2025, Conf42 Machine Learning 2026]
who it's for
machine-learning engineers, data scientists, ml platform engineers, site reliability engineers, security engineers, product managers
topics
Weaviate vector search and HNSW; Hugging Face transformers and Spaces; Amazon Lookout for Vision serverless pipeline; h2o deployment on Kubernetes with Ansible; weak supervision and zero-shot labeling; confident learning for noisy labels; detecting silent failures via drift estimation; visual testing with Eyes SDK and Selenium; Mask R-CNN rooftop segmentation for solar; CI/CD security for ML containers
| title | author | length |
|---|---|---|
| Premiere | Jesus Saldana Gonzalez, Felice Pescatore, Johannes Hotter, Vasco Veloso, Joshua Arvin Lat, Laura Ham, Julien Simon, Mohsin Khan, Andrew Knight, Harika Chebrolu, Aditi Ramaswamy, Anisha Biswaray, Wojtek Kuberski, Karan Singh | 09m |
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… | ||
| Machine learning won't put you out of a job | Jesus Saldana Gonzalez | 14m |
Claim: machine learning displaces repetitive tasks but does not eliminate jobs and already speeds work across domains. Concrete tools include GitHub Copilot… | ||
| AI at the service of Business Agility | Felice Pescatore | 27m |
Describes a pilot called Arinn.ia Digital Agile Master that hooks into operation logs and customer signals, runs ML models in the cloud, and surfaces a small… | ||
| Greenfield vs. Brownfield Data Labeling | Johannes Hotter | 20m |
Distinguishes greenfield (fast prototypes, focus on quantity) from brownfield (improve existing labels, focus on quality) and gives concrete methods: use weak… | ||
| Artificial Intelligence through the ages | Vasco Veloso | 41m |
A fast chronological tour that ties myths to modern failures with concrete examples: Homer’s autonomous ships and Hephaestus’ tripods as early imaginings, real… | ||
| Hacking and Securing ML Systems and Environments | Joshua Arvin Lat | 43m |
Shows how a malicious prebuilt model or container can open a port (example: port 4000) or drop a miner, letting an attacker pivot from a public notebook… | ||
| Build your own search application with Weaviate | Laura Ham | 31m |
Overview of an open-source vector database that stores data as high-dimensional embeddings and runs semantic search, question answering and classification… | ||
| Machine Learning 2.0 with Hugging Face | Julien Simon | 33m |
Hugging Face claims practical transfer learning: a hub with about 2,900 datasets and over 29,000 pre‑trained models in 160+ languages that sees roughly 1… | ||
| Serverless Architecture for Product Defect Detection | Mohsin Khan | 26m |
Shows a working serverless pipeline that uses Amazon Lookout for Vision with S3, API Gateway, Lambda, Step Functions, DynamoDB, Kinesis Firehose, SNS… | ||
| How Visual AI Makes Testing a Breeze | Andrew Knight | 30m |
Demonstrates replacing brittle element checks with visual AI using apple tools Eyes SDK integrated into a Java + Selenium WebDriver suite (add the… | ||
| Using AI to Accelerate the Smart Green Transition | Harika Chebrolu | 11m |
Claims a production-ready deep vision engine that detects rooftops, segments and classifies them into flat / heavy industrial / low-clutter / existing solar… | ||
| Don't Waste Data | Aditi Ramaswamy, Anisha Biswaray | 09m |
Historical USPS event logs get reduced to a transit-time target (earliest vs latest event) and fed into a scikit-learn random forest regression trained on… | ||
| How to detect silent failures in ML models | Wojtek Kuberski | 23m |
Defines two concrete failure modes: data drift (change in input distribution) and concept drift (change in true mapping), and shows concept drift almost always… | ||
| Edge-to-Core Machine Learning and Data Pipeline on k8s | Karan Singh | 25m |
A runnable edge to core pipeline on OpenShift/Kubernetes that deploys LPR and OCR services, local and central Kafka clusters with Kafka MirrorMaker to move… | ||
| nothing matches | ||