1. Posts/

What Have I Been Up To?

·2 mins·

It’s been about seven months since I last posted here (about YAMS), so I wanted to share a quick recap of what I’ve been up to. If you measure that gap in AI progress, we’ve seen an incredible acceleration in both performance and the pace of advancement.

In the time that I’ve been absent, I’ve been working on pushing YAMS forward. I’ve learned a lot of hard-won lessons about building a memory system for AI in C++. In terms of where I want to be, I’m about 75% of the way to getting the project to a place where it’s both usable and beneficial for others.

This year, I plan on experimenting with retrieval optimizations that take advantage of newer NPUs to accelerate embedding generation and add predictive features that tune to usage patterns over time. I’ve reconciled myself to this tooling being a toy project, and any way I can accelerate my own workflow is the direction the project will head in.

On the research side, most of my time has been going into programmable networking and security research. The core idea builds on P4 (a DSL for programming network switches) and deep integrations with graphs and learning systems. The short version: I’m exploring how to use machine learning to accelerate filtering and detection for network attacks, while keeping the system practical for real-world packet processing constraints.

This year, I have a few simple goals for myself. I want to write more, documenting my experience and perspective with the technologies I’m building and interfacing with. I also plan on providing write-ups that cover smaller snippets of my work.

If you are interested in checking out YAMS, you can find out more at these links:

If you’re working in programmable data planes, network security, or applied ML for packet analysis and want to compare notes, feel free to reach out.