Taking the pain out of data labeling


Details
Ivan Lee, CEO and Founder of Datasaur, will talk about the most valuable lessons that he has learned in data labeling from supporting more than 100 NLP projects.
Over the last 4 years, Datasaur has worked with NLP initiatives from across the world, from data science teams at Fortune 100 companies to 3-person startups and academic research groups compiling COVID research reports. This has included decisions at every stage of data labeling, including choosing outsourcing labeling to hiring in-house, automating labeling work and build vs. buy decisions.
This field is rapidly maturing, and there are vastly different approaches to the labeling process. Many people initially view labeling as the unsexy grunt work required to unlock the "fun" work of testing state of the art models and fine tuning parameters. During this talk we'll cover the accumulated best practices for data labeling, show you how to confidently take control of the process, and maybe even have some fun along the way.
Bio:
Ivan Lee is the CEO and Founder of Datasaur. He graduated with a Computer Science B.S. from Stanford University. He dropped out of his Computer Science Master's degree to co-found Loki Studios, a location-based mobile game. Loki was acquired by Yahoo in 2013.
Ivan has spent the last 10 years as a product manager for Machine Learning. Most recently at Apple. Given his own painful experience managing labeling projects, he sought to build the tools he wished existed at the time. He's spent the last 4 years doing just that. Datasaur offers an NLP labeling platform. It is backed by Initialized Capital, YCombinator, and the CTOs of OpenAI and Segment. Datasaur helps companies like Google, Qualtrics and Spotify save 70% of their labeling time and costs and helps them build out higher quality data.
Thank you Diffbot for hosting this event! Please come to 333 Ravenswood Ave in Menlo Park, at the Stanford Research Institute (SRI). Doors will open at 6pm and the talk will start at 7pm. If you can't attend the event in person, you can watch live at: https://zoom.us/s/4252146573
Please note: a government issued photo ID is required by SRI to issue the visitor badge. If you have trouble finding Diffbot after you arrive at SRI, please call 408-722-7670.

Taking the pain out of data labeling