Thank you for sending your enquiry! One of our team members will contact you shortly.
Thank you for sending your booking! One of our team members will contact you shortly.
Duration 14 hours
Course Outline
AI Sovereignty and Local LLM Deployment
- Assessing cloud LLM risks: data retention policies, input usage for training, and jurisdictional concerns.
- Examining Ollama’s architecture: model server components, registry functionality, and OpenAI-compatible API features.
- Comparative analysis with vLLM, llama.cpp, and Text Generation Inference.
- Navigating model licensing terms for Llama, Mistral, Qwen, and Gemma.
Installation and Hardware Configuration
- Setting up Ollama on Linux environments with CUDA and ROCm support.
- Configuring CPU-only fallbacks and optimizing for AVX/AVX2 instruction sets.
- Implementing Docker deployments with persistent volume mappings.
- Establishing multi-GPU configurations and managing VRAM allocation strategies.
Model Management
- Acquiring models from the Ollama registry using commands like ollama pull llama3.
- Importing GGUF format models from HuggingFace and TheBloke repositories.
- Evaluating quantization levels such as Q4_K_M, Q5_K_M, and Q8_0 for performance tradeoffs.
- Managing model switching and understanding concurrent loading limitations.
Custom Modelfiles
- Mastering Modelfile syntax: defining FROM, PARAMETER, SYSTEM, and TEMPLATE elements.
- Adjusting generation parameters including temperature, top_p, and repeat_penalty.
- Crafting system prompts to enforce role-specific behaviors.
- Building and distributing custom models to a local registry.
API Integration
- Utilizing the OpenAI-compatible /v1/chat/completions endpoint.
- Handling streaming responses and enforcing JSON output modes.
- Integrating local models with LangChain, LlamaIndex, and bespoke applications.
- Implementing authentication and rate limiting via reverse proxies.
Performance Optimization
- Sizing context windows and managing KV cache resources.
- Optimizing batch inference and parallel request processing.
- Allocating CPU threads with NUMA topology awareness.
- Monitoring GPU utilization and tracking memory pressure.
Security and Compliance
- Isolating network traffic for model serving endpoints.
- Implementing input filtering and output moderation workflows.
- Maintaining comprehensive audit logs of prompts and completions.
- Verifying model provenance through hash validation.
Requirements
- Proficiency in intermediate Linux administration and container management.
- A foundational grasp of machine learning principles and transformer architectures.
- Working knowledge of RESTful APIs and JSON data formats.
Target Audience
- AI engineers and developers seeking to migrate away from cloud LLM APIs.
- Organizations with stringent data privacy requirements that restrict cloud model usage.
- Government and defense entities necessitating air-gapped language model capabilities.
