I am an ML Infrastructure Engineer focused on the systems that make AI/ML workloads fast, reliable, and scalable in production from accelerator kernels and distributed inference platforms including the CI/CD, networking, and cloud infrastructure they run on.

Currently I work as a Founding AI/ML Infrastructure/Performance Engineer that leads infrastructure at Kernelize, where I build heterogeneous hardware tooling, Triton backends infrastructure, and custom PyTorch runtimes. My core strength is ML infrastructure and operations(MLOps) but I draw on a broad and deep systems background: high-performance networking (DPDK, RDMA, InfiniBand, sub-microsecond latency), Linux/FreeBSD kernel development, hypervisors and virtualization (KVM, Open vSwitch), cloud architecture (AWS), and security engineering in order to design ML platforms that hold up and deliver under heavy production demands.

• ML Infrastructure: Building the Nexus runtime, Triton-Bench benchmarking platform, and Triton/PyTorch plugin inference stack at Kernelize
• MLOps Tooling: Accelerator kernel capture/replay pipelines, hardware profiling with Nsight Compute, and roofline/occupancy analysis for Triton kernels
• Distributed Inference: vLLM, Ray, and custom PyTorch PrivateUse1 backends for heterogeneous hardware
• Compiler Internals: Triton, PTX, MLIR, and torch.compile/Inductor for cross-accelerator kernel portability
• CI/CD for ML: Automated build, benchmarking, and artifact pipelines spanning multiple accelerator targets

Primary — MLOps & ML Infrastructure

• Accelerator kernel tooling, Triton backends, and PyTorch custom runtimes
• Distributed inference platforms (vLLM, Ray) and model serving infrastructure
• Accelerator profiling, roofline modeling, and kernel-level performance engineering
• ML-focused CI/CD, artifact management, and benchmarking pipelines

Supporting Depth — Systems, Cloud, and Networking

• High-Performance Networking: DPDK, VPP, RDMA, InfiniBand, kernel bypass
• Cloud & Virtualization: AWS, KVM, Open vSwitch, Ceph, Docker, Kubernetes, Terraform
• Low-Level Systems: Linux/FreeBSD kernel development, device drivers, hypervisors
• Data Engineering: Pipelines for benchmark data, financial time-series, and fine-tuning datasets
• Security Engineering: Reverse engineering, vulnerability research, secure virtualization

ML Infrastructure     : vLLM, Ray, PyTorch, torch.compile/Inductor, Triton, CUDA, PTX, MLIR
MLOps / CI-CD         : GitHub Actions, Jenkins, Docker, Kubernetes, artifact pipelines
Accelerator Profiling : NVIDIA Nsight, Nsight Compute (ncu), roofline analysis, Flamegraph
Languages             : C/C++, Python, Assembly (x86/ARM), Java, Bash
Cloud & IaC           : AWS, Terraform, Docker, Kubernetes
Virtualization        : KVM, Open vSwitch, Ceph, hypervisor internals
Systems               : Linux, FreeBSD, NVIDIA, DPDK, eBPF
Networking            : RDMA, InfiniBand, TCP/IP, DPDK, VPP
Data Engineering      : Apache Airflow, Spark, Pandas
Security              : IDA Pro, Ghidra, reverse engineering, secure ML pipelines

• M.S., Financial Engineering — WorldQuant University
• B.S., Computer Science — Northern Illinois University