Paschalolo

Systems Engineer · Low-Level Software, Microarchitecture, & High-Performance Infrastructure

Engineering Philosophy

Measure, Don't Guess: Optimization without profiling is just superstition. Microarchitectural bottlenecks—cache misses, branch mispredictions, pipeline stalls—dictate the design.
Zero-Cost Abstractions (When Possible): If an abstraction introduces runtime overhead or hides the underlying hardware cost, discard it.
First Principles: Relying on the standard library or OS primitives is a choice, not a requirement. True understanding comes from building the stack from the metal up.

Languages

C++ (C++23 Modern), C (C11/C23), Assembly (x86_64, ARM64, PTX ISA ) , CUDA ,

Engineering Projects

Kernel-Bypass Storage Engine
C11 | Zero-Copy | Lock-Free
A custom, wait-free storage architecture designed to bypass kernel I/O friction and maximize PCIe throughput.

MicroTrace Debbuger
C | C++ | Assembly
Designed and engineered a zero-dependency, userspace debugging engine in C from first principles, utilizing raw POSIX system calls (ptrace, mmap, waitpid) for dynamic process instrumentation, memory protection manipulation (mprotect), and state-machine execution control.

Custom Dynamic Linker & ELF Parser
C11 | x86_64
Engineered from first principles to resolve symbol tables, manage relocations, and execute dynamic binary loading without libc dependencies.

NUMA-Aware Slab Allocator
C++ | Performance Tuning
A thread-local memory allocation engine optimized for NUMA-topology, effectively eliminating lock contention in high-concurrency workloads.

Profile-Guided Binary Optimizer
Systems Architecture
A binary rewriter that analyzes execution paths to restructure basic blocks, resulting in a 40% improvement in instruction cache efficiency.

Async State Machine Library
C++ | Assembly | SIMD
A stack-based coroutine engine leveraging manual assembly control for ultra-lightweight asynchronous task scheduling.