// writing
Notes from inside the inner loop.
Long-form technical writing from InnerLoop Resources — profiling methodology, latency engineering, and the workforce compliance questions that follow specialized engineers through their careers.
// Workforce compliance
STEM OPT Employment Verification for Performance Tuning Engineers
How performance tuning engineers document STEM OPT employment, build a defensible I-983 training plan, and keep verification records aligned with real systems work.
read →Workforce complianceCPT Guidelines for Low-Level Systems and Performance Engineering
Practical CPT guidance for students entering low-level systems and performance engineering roles: curricular integration, hour structures, and documentation.
read →Workforce complianceH-1B Specialty Occupation Validation for Systems Performance Architects
How to evidence specialty occupation status for systems performance architects: degree nexus, role complexity, and documentation that reflects real engineering depth.
read →Workforce complianceLCA Compliance for High-Comp Systems Optimization Roles
Labor Condition Application compliance for highly compensated systems optimization roles: wage levels, worksite coverage, public access files, and audit readiness.
read →Workforce complianceH-1B Grace Period: Practical Steps Following Unanticipated Layoffs
A calm, sequenced plan for the H-1B grace period after an unexpected layoff: dates, documents, portability filings, and options when the window is short.
read →// Performance engineering
Eliminating Garbage Collection Pauses in Ultra-Low-Latency Java Applications
Techniques for removing garbage collection pauses from latency-critical Java: allocation elimination, collector selection, object pooling, and safepoint analysis.
read →Performance engineeringMemory Profiling and Leak Detection in Complex C++ Microservices
A workflow for finding leaks, fragmentation, and allocator pathologies in production C++ microservices using sanitizers, heap profilers, and allocator telemetry.
read →Performance engineeringOptimizing CPU Cache Locality for High-Throughput Financial Systems
Data layout, false sharing, prefetching, and NUMA placement techniques that raise throughput in order books, risk engines, and market data pipelines.
read →Performance engineeringAsynchronous I/O Programming: Boosting Throughput with Netty and Rust
How event-loop and async runtime design decisions in Netty and Rust determine throughput: batching, backpressure, zero-copy paths, and io_uring.
read →Performance engineeringProfiling Linux Kernel Bottlenecks in Virtualized Cloud Environments
Finding kernel-level bottlenecks on cloud instances: steal time, eBPF tracing, syscall and interrupt overhead, and the limits of virtualized performance counters.
read →