WEB ONLY 15 MODULES INTERMEDIATE

Local LLM Models and Enterprise Agentic Architectures

Learn to select, run and adapt open-weight LLMs on your own infrastructure. Use measured GPU results to compare quantization, fine-tuning and serving, then apply them to enterprise RAG, agents and evaluation. Course material is in English.

Software engineer working with code, diagrams and checks

What's inside

15 modules,
contact us for access.

01
Open-Weight Model Families and Selection Criteria

Compare open-weight model families by quality, licence, language support, context and memory.

Web text + images
LOCKED
02
Running Models Locally: Ollama and llama.cpp

Run the same GGUF file with Ollama and llama.cpp and compare generation speed.

Web text + images
LOCKED
03
GPUs for LLM Inference: Memory, Bandwidth and KV Cache

Estimate memory for weights and KV cache and measure how context length affects inference.

Web text + images
LOCKED
04
Quantization: Size, Speed, Memory and Quality

Compare memory, speed and task quality at different precisions using the same model.

Web text + images
LOCKED
05
Prompting, Structured Output and Tool Calling with Local Models

Design prompts, structured responses and tool calls for local models.

Web text + images
LOCKED
06
Distillation and Supervised Adaptation

Explore how to transfer behaviour to a smaller model and when to use supervised adaptation.

Web text + images
LOCKED
07
Efficient Fine-Tuning with LoRA, QLoRA and Unsloth

Train a QLoRA adapter on an RTX GPU and compare answers before training and after reloading the adapter.

Web text + images
LOCKED
08
Local RAG: Embeddings, Retrieval, Reranking and Citations

Build a local RAG pipeline with embeddings, retrieval, reranking and citations.

Web text + images
LOCKED
09
Repeatable Evaluation: Benchmarks and Evaluation Sets

Define versioned evaluation sets to compare models and adaptations on specific tasks.

Web text + images
LOCKED
10
Serving with vLLM and Monitoring

Measure vLLM throughput and latency at different concurrency levels and monitor GPU usage.

Web text + images
LOCKED
11
Sizing and Economics of a Local Solution

Size a local solution and compare hardware purchase, GPU rental and per-token costs.

Web text + images
LOCKED
12
Security, Licenses and Data Protection

Examine model licences, model provenance and data protection for a local deployment.

Web text + images
LOCKED
13
Local Coding Agents: OpenCode and Ollama

Use OpenCode with Ollama and compare local models on the same coding task.

Web text + images
LOCKED
14
Enterprise Agentic Architectures on Local Models

Examine the components of an enterprise agent architecture on private infrastructure.

Web text + images
LOCKED
15
Local Judges and Decision Models

Evaluate local models for classification and decision tasks, comparing calibration, cascades and human review.

Web text + images
LOCKED