AI Engineering Bootcamp: From Proof of Concept to Deployment

AI Engineering Bootcamp Curriculum

• 🎯 The AI engineering bootcamp is designed for individuals at an intermediate to advanced level, aiming to elevate them to an advanced or advanced-plus proficiency.
• 🔑 Prerequisites include existing knowledge of Python, PyTorch, scikit-learn, and Pandas.
• 💡 Familiarity with LLM API calls and some cloud experience is also expected, as these are not covered in depth during the bootcamp.

Core Components of AI Engineering

• 🧠 Successful AI engineering requires a blend of the science of AI (building proofs of concept) and deployment into real-world environments.
• 🛠️ The first four weeks focus on the science, including LLM selection, RAG augmentation, and the use of agents to solve specific business problems.
• 🚀 The latter four weeks concentrate on deployment, covering how to take a proof of concept from a Jupyter notebook to a cloud environment like AWS, Azure, or GCP.

Deployment Considerations

• ☁️ Key deployment aspects include making AI systems secure, efficient, cost-effective, reliable, and scalable to handle varying user loads.
• ⚖️ Decisions on architecture, such as serverless vs. server-based, involve trade-offs in responsiveness and latency.
• 📈 The bootcamp emphasizes understanding these critical constraints for real-world business applications.

Bootcamp Instructors and Approach

• 🌟 The bootcamp features two instructors: one focusing on the possibilities of AI and another on practical deployment realities.
• 🤝 The first instructor, Ed Donner, introduces AI capabilities, creating a vision of what's possible.
• ☁️ The second instructor, Sam Baston, an expert in cloud and LLM deployments, grounds these possibilities in the practicalities of real-world implementation, managing expectations for what can be deployed.