martinuke0's Blog

Illustration of memory arenas and thread caches in a multi‑core server.

Deep Dive into jemalloc Arenas and Thread Caches: Architecture, Scalability, and Memory Management Patterns

A technical walkthrough of jemalloc’s arena and thread‑cache subsystems, showing how they achieve low contention and high throughput in real‑world services.

Diagram of tri‑color marking stages overlaid on a memory heap.

Implementing Concurrent Garbage Collection: Deep Dive into Tri-Color Marking for Low-Latency Memory Management

A practical guide to building a concurrent garbage collector using tri‑color marking, covering core invariants, integration with JVM and Go runtimes, and real‑world performance tuning.

Implementing WebGPU-Accelerated Quantization for Local Llama Inference: A Deep Dive into Browser-Based Performance

A step‑by‑step guide that shows engineers how to combine WebGPU with weight quantization to run Llama locally, complete with code snippets and production‑grade patterns.

Diagram of two buckets—one with tokens spilling out, the other leaking water—illustrating rate‑limiting concepts.

Architecting Production Rate Limiters: A Deep Dive into Token Bucket vs. Leaky Bucket Algorithms

A production‑focused guide that compares token bucket and leaky bucket rate limiters, showing how to choose, implement, and observe them at scale.

Diagram of a Celery worker pool processing tasks from a broker.

Architecting Scalable Python Applications: Using Celery as a Distributed Task Queue for Production Pipelines

A deep dive into using Celery as a distributed task queue for scalable Python applications, with concrete architecture diagrams, code samples, and operational best practices.