Every slash command, CLAUDE.md, skills, hooks, subagents, MCP, plugins and CI: agentic development from A to Z.

Search and graphs in practice: 200,000 indexed articles, Kibana, OpenSearch, Cypher and recommendation, in a tested Docker kit.

Pipelines, feature stores, model registries, monitoring and rollback.

Ground your LLM in real, up-to-date data, without retraining.

LoRA, QLoRA and full fine-tuning: pick the right one and do it well.

Architectures, training, evaluation and safe deployment of LLMs.

Autonomous agents that plan, use tools and act, with sane guardrails.

Compose LLMs, tools and memory into real applications.

Run capable open-source models on your own hardware, offline and private.

Structured prompting techniques that outperform intuition every time.