Hey there!
I’m Josh, a systems software engineer at Arm in Austin, Texas. Most of my work sits close to the machine: performance analysis, scheduling, and understanding how software behaves on real hardware under real load.
What I Do
Lately I’ve been focused on ML workloads on Arm platforms, especially what happens when multiple models need to share the same system. That means profiling bottlenecks, reasoning about contention, and building better scheduling strategies so the hardware is used well instead of just used hard.
I tend to enjoy the kinds of problems where the right answer is not more abstraction, but better understanding. The work I like most usually involves tracing behavior through the stack, measuring before guessing, and making systems faster, more predictable, and easier to reason about.
The areas I keep coming back to are:
- Systems software: performance analysis, scheduling, resource management
- Machine learning systems: inference performance, concurrent workloads, efficient execution
- Embedded and edge computing: hardware-aware optimization, constrained systems, deployment tradeoffs
Research & Publications
I’ve worked on a few research projects around edge computing, memory systems, and ML infrastructure:
- “ML-ACE: Machine Learning Admission Control at the Edge” - Exploring intelligent resource management for edge ML workloads
- “UDON: A case for offloading to general purpose compute on CXL memory” - Investigating memory architectures for high-performance computing
- “SMARTER: Experiences with Cloud Native on the Edge” - Bridging cloud technologies with edge computing constraints
Background
I studied electrical and computer engineering at the University of Texas at Austin, where I was drawn pretty early toward low-level systems work. A lot of the projects I enjoyed most involved embedded systems, performance constraints, and the interaction between software decisions and hardware behavior.
Some of the projects that still feel representative of my interests:
- Embedded Face Tracking Camera - Real-time computer vision on resource-constrained hardware
- HookedIn - Class scheduling application with smart optimization algorithms
- Embedded Flappy Bird - Game development on microcontrollers
Outside Work
Outside of work, I spend a lot of time playing tennis and taking advantage of Austin’s food scene. Tex-Mex remains a reliable source of energy for both debugging and general decision-making.
Let’s Connect
If you’re interested in systems software, performance engineering, ML infrastructure, or embedded systems, feel free to reach out through the links on my homepage or email me directly at josh.s.minor@gmail.com.