Résumé/CV: I can email it to anyone who reaches out to me
GitHub: <a href="https://github.com/JeremiahM37" rel="nofollow">https://github.com/JeremiahM37
Email: jeremiahmackey91@gmail.com
Recent CS grad (Montana State, May 2026). I spent the last year writing production C for wolfSSL's embedded TLS library, including post-quantum and FIPS work, and built an internal LLM-based vulnerability finder there with one other engineer.
On my own time I work on inference and systems: I ported SGLang to AMD's Strix Halo, which nothing supported, and found through profiling that decode was memory-bandwidth bound, ending at 4.76× throughput under concurrency. I've had patches merged into Tailscale and AMD's aiter kernel library, and I run my own agents against a self-hosted cluster with a nightly eval harness.
Looking for: inference/ML infrastructure, systems, security, embedded, or agent/eval work. I'm best at embedded and systems level stuff.
jeremiahm37 · · focus · HN ↗
Willing to relocate: Yes (NYC, Seattle, Bay Area preferred, but willing to relocate anywhere)
Technologies: C, C++, Python, Go, TypeScript/React, ROCm/HIP, SGLang, llama.cpp, PyTorch, Linux, Docker, Kubernetes, Terraform, Ansible, FreeRTOS, Zephyr
Résumé/CV: I can email it to anyone who reaches out to me GitHub: <a href="https://github.com/JeremiahM37" rel="nofollow">https://github.com/JeremiahM37
Email: jeremiahmackey91@gmail.com
Recent CS grad (Montana State, May 2026). I spent the last year writing production C for wolfSSL's embedded TLS library, including post-quantum and FIPS work, and built an internal LLM-based vulnerability finder there with one other engineer.
On my own time I work on inference and systems: I ported SGLang to AMD's Strix Halo, which nothing supported, and found through profiling that decode was memory-bandwidth bound, ending at 4.76× throughput under concurrency. I've had patches merged into Tailscale and AMD's aiter kernel library, and I run my own agents against a self-hosted cluster with a nightly eval harness.
Looking for: inference/ML infrastructure, systems, security, embedded, or agent/eval work. I'm best at embedded and systems level stuff.