AI Engineering
A collection of notes on AI engineering and related topics. Includes tools and resources for AI engineering.
Introduction to AI Engineering
AI Engineering is an emerging discipline focused on building, deploying, and maintaining AI systems at scale. Unlike traditional machine learning approaches, modern AI engineering primarily leverages pre-trained foundation models rather than training custom models from scratch.
What is AI Engineering?
AI Engineering bridges the gap between research-oriented AI development and production-ready systems. It involves:
- Adapting and fine-tuning pre-trained foundation models for specific use cases
- Engineering effective prompts and interactions with AI models
- Building applications and systems that integrate with foundation models
- Ensuring reliability, scalability, and security in AI-powered applications
- Creating seamless user experiences around AI capabilities
Core Components
- Foundation Model Utilization: Leveraging large pre-trained models (LLMs, diffusion models) rather than training from scratch
- Prompt Engineering: Designing effective prompts to achieve desired outcomes from foundation models
- Retrieval Augmented Generation (RAG): Enhancing model outputs with relevant external information
- AI Integration: Connecting foundation models with existing software, APIs, and data sources
- Responsible AI: Ensuring ethical use, safety guardrails, and mitigating bias in AI applications
Current Trends
- Fine-tuning foundation models on domain-specific data
- Building AI agents with specialized capabilities
- Developing tools that simplify AI integration into workflows
- Creating multi-modal applications that combine text, image, and other data types
Resources
Tools
- LLM chats
- AI agents
- Code analysis
- gitingest: A tool to ingest and analyze git repositories. Turn any Git repository into a simple text digest of its codebase. This is useful for feeding a codebase into any LLM.
- Books
Articles
- My LLM codegen workflow atm. Great article by Harper Reed on his LLM codegen workflow.
Written by Jaime González García , dad, husband, software engineer, ux designer, amateur pixel artist, tinkerer and master of the arcane arts. You can also find him on Twitter jabbering about random stuff.