Monika Sharma is an AI researcher focused on convolutional neural networks and camera stream processing. She leads the physical intelligence division at Primine.
In this recorded podcast, Monika Sharma breaks down the computer vision technologies that enable accurate face identification.
We look at the math behind embedding models and face vector distance algorithms, explaining how a camera matches a face against database images in fractions of a second.
Monika also discusses the moral responsibilities of developers, detailing how Primine ensures that our physical security systems operate without storing personal data, complying with the EU AI Act.
Sarah Jenkins: Welcome everyone to another episode of The Primine Podcast. Today we are talking about engineering challenges.
Monika Sharma: Thanks Sarah. When looking at this problem, many teams jump straight into the application layer without realizing the structural bottlenecks underneath.
Sarah Jenkins: That's a great point. How does that translate into long-term cloud scalability and efficiency?
Monika Sharma: Well, it's about minimizing network latency, designing standard schemas, and choosing the right compute locations.
Register to obtain standard text transcription sheets and lecture files.
Get Resources Folders