Over the holidays I intend to build (or fork NetNewsWire TBH) an RSS reader app that uses the Apple on-device models (or BYOK) to summarize and prioritize articles - my very own personalized algorithmic feed. Curious to see how it turns out.
My takeaway from the demo is less that "it's different each time", but more a "it can be different for different users and their styles of operating" - a poweruser can now see a different Settings UI than a basic user, and it can be generated realtime based on the persona context of the user.
Example use case (chosen specifically for tech): An IDE UI that starts basic, and exposes functionality over time as the human developer's skills grow.
Microsoft | Cybersecurity | Seattle Area ONSITE (no in-office requirement) | Senior Security Engineer
The Azure Security DevSec team is looking for a Senior Security Engineer to join us in securing services in the Microsoft Cloud. We work with service teams across Microsoft to evaluate and improve their security posture in the cloud, continuously moving the needle across multiple dimensions of their design, development and deployment security. In this role, you will collaborate with teams across Microsoft to dive deep and evaluate the risk posed by and to a given service, recommend existing mitigations, innovate on mitigations for new issue classes, contributing to the feedback loop that can help achieve these outcomes at scale.
I’ve built something for a solution that takes you most of the way there, using Semgrep’s SARIF output and prompted LLMs to help prioritize triage.
We’ve used this for the past year at Microsoft to help prioritize the “most likely interesting” 5% of a large set of results for human triage. It works quite well…
As part of solving a code review exercise in a large inventory of code, I resorted to using Static Analysis + LLMs to capture and summarize, analyze code. The approach yielded useful results, and made me rethink SAST rule patterns.
I just read the entire Chapter 3 of your O'Reilly book "TinyML" and LOVED how you've made the big-picture of ML training and inference approachable.
I will likely not read any further (since this isn't my area of expertise), but am grateful for the knowledge gained from that chapter. Thank you for putting in the time and energy in sharing this. Much appreciated!
Microsoft has a “Democracy Forward” team (previously called “Defending Democracy”) that aims to protect government officials and systems from adversarial state actors. It’s been ongoing for a few years now.
Given their track record, I'd trust Microsoft approximately 0% to secure my critical/sensitive systems. The funny thing is that the U.S. government does, in fact, trust them.
Check out “Fenix” (3rd party client on Apple platforms) - it mimics the multi-column view, but is more flexible; a list can be a Twitter list, a search query etc.
I haven't used Fenix before (it does look nice!), but I can definitely do all of the above in tweetdeck and did so in the past before raising my walls by following only specific users.
Anyone else already tried something similar?