OpenRoot: supporting the growth of open-source projectsThe mission There are several reasons to why I started OpenRoot. For one is because I feel and see the struggles of independent open-source developers. Im frustrated that many do not get the recogniti1d ago · 4 min readblog.kaelux.dev
Fine-Tuning Is Often the Wrong First Move: Introducing HarneloopWhen an agent repeatedly fails at a narrow task, changing model weights is often not the most direct fix. The failure may come from missing context, weak tools, incomplete retrieval, poor validation, 21MJul 23 · 4 min readblog.kaelux.dev
Kaelux: Engineering the Future of Intelligent InfrastructurePublished by Kaelux AI Engineering — a global agency building custom LLM systems, RAG pipelines, and intelligent automation for businesses. The Problem with One-Size-Fits-All AI Frontier models are i1Apr 6 · 4 min readengineering.kaelux.dev
What We Learned From Analyzing 28,000 Production AI System PromptsOver the last few months developing PromptTriage, we've collected and analyzed over 28,000 production system prompts. Most are bloated, contradictory, and actively hurt reasoning quality. 📉 Anti-Pat1Mar 24 · 2 min readblog.kaelux.dev
AI Format Wars: Does the Shape of Your Prompt Matter? (1,080 Evals Later)AI Format Wars: Does the Shape of Your Prompt Matter? (1,080 Evals Later) We spend hours tweaking the words in our prompts, but how much thought do we give to the structure? If you ask an AI to return1Mar 22 · 5 min readblog.kaelux.dev
