About Karthik and LatentMesh
I'm an engineering leader at Microsoft with 19+ years of experience building systems at scale. LatentMesh is where I write about AI agents, evaluation, safety, privacy, compliance, and governance from the perspective of someone building these systems in practice.
Background
My work sits at the intersection of large-scale systems, AI evaluation, privacy, and governance. At Microsoft, I lead engineering in Purview Data Lifecycle Management, building services for retention, compliance, and governance across large enterprise environments.
Over the past several years, my focus has increasingly moved toward AI: agent systems, evaluation pipelines, retrieval architectures, and compliance automation. That background shapes how I think about AI. Not just as a model problem, but as a systems problem.
What I write about
Most writing about AI focuses on what models can do. I'm more interested in what happens after deployment: the failures nobody planned for, the evaluations that looked good in staging and fell apart in production, the compliance requirements that arrive as legal text and somehow need to become running code.
These are engineering problems, not policy abstractions. That's the gap I write into.
Why LatentMesh exists
A lot of writing about AI stays at the level of model capability or product excitement. I am more interested in the engineering reality underneath it: what breaks, what must be measured, what must be owned, and what must be proved.
LatentMesh exists to make that layer legible.
Connect
If you're working on AI evaluation, compliance, governance, or production agent systems, I'd be glad to connect.
Disclaimer: The views expressed on this website are my own and do not represent the views of my employer.