Building Applications
Local RAG, structured output, function calling and agents on top of your own model.
0 of 7 complete
- ▶ Up next16 min+50 XPLocal RAG from scratchChunk, embed, retrieve and answer over your own documents, entirely offline: embedding model conventions, a…
- Pro12 min+50 XPStructured output: JSON schemas and grammarsHow constrained decoding guarantees valid JSON from small models, how to use JSON Schema and GBNF in llama.cpp…
- Pro14 min+50 XPFunction calling (tools) with local modelsHow tool calling actually works under the hood, the full call loop with validation, parallel calls and errors…
- Pro14 min+50 XPBuilding local agents that don't fall apartDesign agent loops that work with 7 to 14B models: error compounding maths, state and context budgeting, a complete…
- Pro20 min+50 XPRAG in depth: chunking, hybrid search, fusion, reranking and permissionsEverything between 'we have documents' and 'the model answers from them reliably': chunking strategies with a…
- Pro18 min+50 XPAgents: loops, memory, planning and the Model Context ProtocolHow tool-using agents work as a controlled loop, state and memory, planning versus reacting, limits and failure…
- Pro14 min+50 XPPrompting that works on small local modelsWhy small models need different prompting: explicit structure, few-shot examples, constraints, decomposition, output…