Botmonster Tech
Run Google Gemma 4 26B MoE on a budget 8GB VRAM GPU using aggressive quantization, GPU-CPU layer offloading, and tensor parallelism. Full setup guide.
AI | page 21 of 26
Hands-on guides to LLMs, agents, prompt engineering, and the AI tools Botmonster runs every day for real work, not demos.
AI coding agents are insider threats: prompt injection, MCP exploits, and supply chain attacks
Study of 78 AI coding agents including Claude Code, Copilot, and Cursor: all fall to prompt injection, which succeeds 85% of the time via MCP exploits.
Running Gemma 4 locally with Ollama: all four model sizes compared
Google's Gemma 4 spans 2.3B to 31B parameters. Compare hardware needs, speed, and quality across all variants when running locally through Ollama.
Three tiers of AI pair programming: from autocomplete to autonomous overnight agents
Three tiers of AI development: inline completions for flow, agent sprints for features, and overnight batch runs. Route each task to the right one.
Self-hosted AI search: combine SearXNG and a local RAG pipeline
Build a private AI search engine with SearXNG and RAG. Runs on single machine with 12 GB VRAM, delivering cited answers without external queries.
Claude Code skills ecosystem: 1,340+ installable agent skills for AI coding assistants
Explore Claude Code's 1,340+ agentic skills ecosystem: major repositories, top skills by category, building custom skills, and future directions.