Context engineering: what fits and what gets dropped
Originally published on Loop & Retry — field notes on building LLM agents that survive production. "Context engineering" showed up this year as…
DevOps
⚑ Report a ProblemLatest DevOps news from Tech News
Context Window Management for Long-Running AI Agents: 4 Patterns That Work
Originally published on robatdasorvi.com The agent had been running for four hours and it was arguing with itself about a file it had already fixed.…
Multi-agent work in three spoonfuls: what worked for me and what did not
This started as a practical annoyance: long tasks interrupted by quota limits, conversations that became too heavy, and reviews that required copyin…
I stopped writing prompts. I write the repo.
Originally published on indiecore.net . For about a year I got better at asking. Longer prompts, worked examples, careful phrasing, a private collec…
The Glass Box: Inside OpenAI's Computer History & Agentic AI
🎙️ Short on time? Explore the 10-Min Interactive Visual Deck first ➔ Every major breakthrough in software automation begins by removing a layer of h…
OpenCode Needs Memory, Not Another Background Pipeline
I kept explaining the same project facts to new OpenCode sessions. Not the durable rules that already belonged in AGENTS.md . Those were present. Th…
Context window sizing for fine-tuning: how long should your training examples be?
Originally published on Loop & Retry — field notes on building LLM agents that survive production. Most fine-tuning guides answer "how many exam…
The End of "One-Shot AI": Why Context Engineering Is Replacing Prompt Engineering
Most "prompt engineering" advice circulating today is already obsolete for anyone building production-grade AI. Granular phrasing matters for simple…
The Multi-Agent Memory Problem: Why Retrieval-Time Inference Breaks Down at Scale
Published by the Alchemyst AI engineering team. We built Alchemyst AI, the context layer described in this post. This is not a neutral third-party r…
Load late, load little: just-in-time context for conversation history
Most agents drag their entire past into every turn. A better default: keep a thin index of what was said hot, and fetch only the few turns you actua…
The Website Is Not the Product
TL;DR Anthropic's crawler now retrieves 60,000 pages per visitor sent back — up from 6,000:1 six months ago. The website as distribution channel is…
Context Engineering: Building More Reliable LLM Systems in Production
Context Engineering: Building More Reliable LLM Systems in Production In LLM-based systems, performance is often driven less by model size and more…