TL;DR I try to keep my eyes on the AI agents. I gave one too much rope once, and the kind of mess it made while I wasn't watching is something I'd rather not retell. Which is why I needed 5 monitors. To run 5 agents in parallel, 5 VSCode windows have to live in one field of view. Physical monitors hit a wall. No desk fits five; even my viewing angle gives out before the desk does. So I strappe
Modern yazılım geliştirme ekosisteminde altyapının kod olarak yönetilmesi hız ve ölçeklenebilirlik açısından devrim yaratırken GitOps yaklaşımı bu süreci merkezi bir doğruluk kaynağına bağlamaktadır. Ancak tüm yapılandırma detaylarının tek bir platformda toplanması kritik siber güvenlik risklerini de beraberinde getirmektedir. Nesil Teknoloji olarak TSE A Sınıfı sızma testi yetkimizle endüstriyel
Comments
A some time ago I shipped a desktop app to generate LLM fine-tuning datasets. It worked: my Qwen2.5-Coder-7B fine-tune jumped from 55.5% → 72.3% on HumanEval. Whole pipeline ran on OpenRouter — pick a model, click Generate, get JSONL. v1.0.3-beta ships multi-provider LLM support — Ollama, LM Studio, llama.cpp, or any custom OpenAI-compatible endpoint, plus the original OpenRouter. Mix and match: g
Hello Developers! 👋 Most developers today pick a side: Let’s talk about combining C++ and JavaScript—the ultimate hybrid stack for high-performance applications. 👇 1. The Core Engine (C++) ⚙️ 2. The Browser Bridge (WebAssembly) 🌉 3. The Cinematic Experience (Vanilla JS + UI/UX) ✨ The Takeaway 🎯 Keep optimizing, keep building! 💻✨ ~ Ujjwal Sharma | @stackbyujjwal About the Author 👨💻 Ujjwal
A beautiful personal tribute to the practice of programming, interrupted by the switch to LLMs. Comments
I built a Vamana-based vector search engine in C++ called sembed-engine. Recently I made a pull request that sped up queries by 16x and builds by 9x. The algorithm stayed exactly the same. The recall stayed at 1.0. The number of visited nodes did not change. The speedup came from data layout. The original code stored vectors as separate objects pointed to by shared_ptr: struct Record { int64_t
The first time I implemented Vamana from the DiskANN paper, my approximate nearest neighbor index was slower than brute force. On tiny test fixtures, brute force took 0.27 ms per query. My Vamana implementation took 22.98 ms. That sounds absurd. ANN exists to skip work. The problem was not the algorithm. It was how I mapped the paper's abstractions to actual data structures. The DiskANN pseudocode