Tagged: Llm
13 posts
EdgeHome Harness: Why MiniCPM5-1B Never Touches Your Devices
MiniCPM5-1B runs a smart home because a Rust harness treats it as untrusted input. What EdgeHome Harness owns, why frontier agents need the same.
Your AI Agent Makes Four Bad Decisions a Smarter Model Won't Fix
AI agents fail decisions for the same four reasons people do. The WRAP framework, encoded as a system-prompt gate, fixes what a bigger model can't.
AI Agent Best Practices: Trust Your Own Results Before Google
Your AI agent reaches for googled best practices before your own proven fixes. Wire a trust order into your CLAUDE.md and agent loop instead.
Build the Harness Once With Your Best Model. Run It on a Cheap One.
Agents forget and good ones cost. The fix is not a better model. Put the goal in deterministic scripts and run a cheap model against them.
Claude Opus 4.8 Is Out. The Number I Care About Isn't on the Benchmark Chart.
Opus 4.8 shipped May 28. For unattended cron agents, the upgrades that matter are not the benchmark scores. A use-case breakdown from real builds.
Context Engineering Is Just File Naming
Context engineering sounds new. It is the file-naming hygiene developers always had, load-bearing now because LLMs read what you point them at.
Enterprise AI PII Redaction System for Sensitive Documents
Self-hosted German PII redactor for SAP prod→dev copies. Plugs in after TDMS/Delphix/Informatica to cover free-text NOTES columns, unclassified Z-tables, and OCR'd attachments. DSGVO-konform, Apache 2.0, runs on a single consumer GPU.
How to Choose an LLM for Production: 7 Criteria That Matter
How to choose an LLM for production workloads. 7 selection criteria, a decision tree, an evaluation process, and a requirements checklist from real deployments. Download the free AI Automation Checklist.
Self-Hosted LLM vs API Cost: Break-Even Analysis (2026)
Self-hosted LLM vs Claude API in September 2026: current GPU and token prices, recomputed break-even, and when the API still wins.
LLM API Comparison 2026: Claude, OpenAI, Gemini for Production
LLM API comparison for production, September 2026: Claude, GPT, Gemini, Mistral on cost, EU data residency, uptime SLAs and lock-in for IT leaders.
LLM API Cost Comparison 2026: Framework, Not a Stale Table
LLM API cost comparison for 2026. Model your real workload costs with prompt caching, output tokens, reasoning, and batch API factored in. Download the free AI Automation Checklist.
Self-Hosted LLM on Kubernetes: A Production vLLM Deployment
Complete self-hosted LLM Kubernetes guide. Deploy vLLM on GPU nodes with manifests, HPA, monitoring, and cost modeling. Practitioner notes included. Download the free AI Automation Checklist.
RAG Pipeline Tutorial: Build a Production Document Q&A System with Qdrant and Claude
End-to-end RAG pipeline tutorial. Qdrant + Claude Sonnet 4.6 + local embeddings. Real code for chunking, retrieval, augmentation, and citation-grounded answers. Download the free AI Automation Checklist.