https://
future.snorkel.ai conf. is 4 weeks away, June 7-8! We are bringing together the best & brightest from Fortune 500, research, & AI-first companies: @CSETGeorgetown
, @comcast
, @cvshealth
, @Snorkel AI, & @Gretel_ai to share how to build better experiences & products. Join us
@snorkelai
-
Snorkel AI Conference June 7-8: Industry Leaders Share AI Insights
By
–
-
Wayfair Uses LLMs Data-Centric ML Image Tagging
By
–
Join Archana & Vinny at http://
future.snorkel.ai conf. and learn how @Wayfair utilizes #ML to make online furniture shopping fun. They will talk about using #LLMS & data-centric approaches to automate image tagging & improve collaboration with domain experts for faster development -
Lambda Labs GPU Workstation Competition for Deep Learning Research
By
–
Want a GPU workstation for deep learning with 2 @NVIDIA RTX 4090 GPUs with #Ubuntu, #TensorFlow, #PyTorch, & #CUDA pre-installed?Showcase your research on #LLM, #genAI, #promptengineeering, #datacentricAI to win $15,000 in prizes from @lambda_labs & get featured on
-
HackAPrompt Competition: Learn and Master LLM Prompting
By
–
Prompting is a quick and easy way to provide human feedback to #LLMs. We are excited to partner with http://
LearnPrompting.org for the HackAPrompt competition. Learn, share, hack, and win -
Compositional Soft Prompting Improves Foundation Model Understanding
By
–
Compositional Soft Prompting (CSP) is a clever technique from @stevebach & team at @BrownUniversity that helps improve how well foundation models like CLIP understand and work with combinations of words and images. It teaches models to recognize various combinations of attributes
-
Autoscaling compute clusters for parallel job execution
By
–
5) Enabling autoscaling of the compute cluster to execute more jobs in parallel during periods of high activity and save cost during periods of low activity.
-
Ray Parallelization and Job Checkpointing for Distributed AI
By
–
3) Leveraging Ray to parallelize work at large scale across multiple worker pods in the cluster to achieve performance benchmarks
4) Implementing job checkpointing ensures that jobs always run to completion and users see minimal interruption. -
Containerization and Kubernetes scaling for ML workloads
By
–
1) Containerization provides consistent, reproducible behavior required for ML workloads.
2) Scaling gets first-class support in Kubernetes, unlocking scaling for training and experimentation. -
Kubernetes Lessons Learned Running ML at Snorkel
By
–
Snorkel tackles heavy ML workloads for some of the world's largest organizations. We have built our infrastructure on Kubernetes—even though it wasn't designed for ML! Here are a few lessons learned from @wheeliamhuang
, Tech Lead Manager at Snorkel https://
snorkel.ai/kubernetes-les
sons-learned-at-snorkel-ai/
… -
Data labeling bottlenecks overcome with foundation models automation
By
–
Data is the new moat, differentiating businesses in the AI age! You can now harness foundation models & LLMs for powerful automation by overcoming bottlenecks in data labeling. @ajratner dives deeper with @ABridgwater
, @ComputerWeekly https://
computerweekly.com/blog/CW-Develo
per-Network/Auto-tech-series-Snorkel-AI-Data-is-the-new-input-for-business-process-automation
…