Link below! I'm #hiring a Senior #SWE (L5) at Google Los Angeles to work on the ☣ #Ads Content Safety ☣ team. You will help build cutting edge #MachineLearning and Serving infrastructure for enforcing Google Ads platform Policy and Creative Quality. We train and deploy large

Here's an interesting example of how #LLMs can still stumble on explaining concepts like Market Maker #gamma hedging techniques. It's not that it doesn't understand the strategy MMs use, it's actually a pure mathematical error trying to explain how they hedge with ES futures yet

10 years today at Google. What a ride. Learning from the best and watching AI evolve while being at the front lines of it all. To all my friends, colleagues, and family....thank you for the best years of my life. For now, we continue the journey and see what lies ahead. Cheers!

❗️🚨Dr Steven Quay this tweet, without proper explanation of the real data, can lead to inaccurate claims or applauds for the US medical system which has done a bit poor in the field of Prostate Cancer detection. 18% of all men who do a yearly PSA test get a positive diagnosis. Another

I'm not here to defend or support any claims that the diagnosis was covered up. I'm simply here to send a message to the public that you should do your own homework when it comes to detection rates of cancers like Prostate. Most especially make sure to double check all tests with

Just ~2 months in and AI Mode already has over 100 million monthly active users in the U.S. and India, and it's great to hear the awesome response from our users!! Now, we’re excited to ship AI Mode, our most powerful AI search, to the UK! Starting to roll out today. 🇬🇧

Scaling agents to gather both synthetic and human quality labeled data is going to be a big upgrade for many companies out there wishing they had the capability to fine tune models for their use cases, but lack expertise and producing/finding similar examples to their current

🚀 Let's go!

Andy Masley There are major uncorrected factual errors in the heavily cited paper Strubell et al. paper (5800+ citations) about the CO2e emissions of language model training workloads, as well, including confusing one-time costs for neural architecture search (NAS) vs. training a language

Imagine knowing all the cards, and who holds them.

Jaana is at the forefront of agent orchestration. I'm excited to partner with her internally as my team works with her to flesh out agentic platforms for 1P Google use cases! Jaana Dogan ヤナ ドガン we haven't met yet, since I just joined AIPG in December, but I reviewed your doc already

Skew pricing, tail risk and VVIX are rising "more consistently than the VIX" recently - Bloomberg, 2/1/26 Each of these things tells you something about the nature of the options positions being managed beneath the vol surface Take skew, for example >>

Class is in session, listen up.