Ben Herzberg’s Post

This is *exactly* why context is so important in data security. Just got an email from Youtube, about a video we published that "violated" the community guidelines and was removed. The reason is the "policy on personally identifiable information". The video itself? it shows a demo of how to protect a PII dataset in Satori (obviously with a synthetic dataset). But this goes to show how important it is to understand the context when protecting data: -> What is the reason a user needs access to sensitive data? -> For how long? in what terms? can the data be de-sensitized? -> What compensating factors are there? ... and so on.

In today's data-driven world, safeguarding sensitive production data is crucial. Granting permanent access to engineers and analysts can increase the risk of data breaches and compliance violations. Temporary credentials offer a safer solution: 🛡️ Limited Access: Grant access on a need-to-know basis, reducing risks. 👮 Enhanced Security: Prevent over-privileging and minimize data exposure. ⚡ Streamlined Operations: Automate provisioning for efficiency and compliance. Learn how Satori's temporary credentials can protect your data while empowering your teams. Link to blog post in first comment...

After some stressful months (which also canceled my PTO to India :/), I grabbed the chance of having two consecutive events in England (Databricks Data+AI London last week and Big Data LDN this week), and over the long weekend I scratched 🏴󠁧󠁢󠁷󠁬󠁳󠁿 Wales 🏴󠁧󠁢󠁷󠁬󠁳󠁿 off my bucket list! By chance (or perhaps not by chance? Who knows??), after a long day of hiking in the Welsh coastline, attempting to find my way back to the car before the sun is down... Of all things... I found a small Ganesha looking at the sea! Was it waiting there to remove some of my obstacles? Was it waiting there to comfort me after not going to India over the summer? Or perhaps it just happened to be there, left for the joy of hikers like me.

It's that time of year again! I will be at Bigdata London next month, and if you're in data, you should be too.

Why are we training models if not to win games 🂡🂫? In this post: https://lnkd.in/d4_dBxrT, Yoray Herzberg trains a DQN in Rust to play Blackjack. Will it give an edge over the dealer?

A couple of weeks ago at CDOIQ Program I had the privilege of co-presenting with Arun K. Buduri (from Innovaccer) about D&A governance for AI in Healthcare. If you weren't in the room, here's a link to the recording of our session. Also, thanks Arun K. Buduri always a pleasure meeting with you! https://lnkd.in/dmrReKye Tamar Ascher Naim Satori ★ Ari Oster Eldad Chai Yoav Cohen Tomer Shani Oleg Toubenshlak Aneesh Sharma Rohit Ranjan Seth Cutler Evgeny A

When we use random values, we tend to forget that they're mostly Pseudo-random. But can you really "guess" the next random number? And if so - how? And what are the implications? Read the full article by Yoray Herzberg. https://lnkd.in/d4rKNwBJ

Yesterday in #rustIL , Yoray Herzberg (DD: my son 😻 :D) gave a talk about how he implemented Neural Networks in Rust. Great stuff if I may add - explained the algorithm behind, and some of the code implementing it. Also, great talk by Ron Zigelman about creating Microservices, Message Brokers, and Protobuf with Rust! Thanks Gábor Szabó for organizing this event in the Microsoft for Startups hub in TLV! Tagging some awesome rustafarians :) David Levy Nahum Zuf Pilosof Tomer Shani eyal leshem Alon Ludmer jacob kilimnik Aviv David Evgeny A Boaz Sade Omer Shtivi shlomit danin Stan Chlebicki Dawid Karolewski Joe Rozner Maciej Skrzypkowski Alex Yaroslavsky

Struggling with complex DAX expressions for RLS in Power BI? Struggling with keeping them up to date with your needs? There might be a better way! Check out Guy Raz's latest blog on how Satori simplifies RLS with ABAC #PowerBI #RLS #ABAC https://lnkd.in/dEMjmPy6

EU AI Act is going to gradually transform a lot around how enterprises use AI. Read Lisa Cave LevyLevy's article to get a glimpse of some of the upcoming challenges. https://lnkd.in/dtkiYCDC