Specialty Track: Technical Builders
The AI Infrastructure Path
Who This Track Is For
Aspiring data scientists, IT professionals, developers, and anyone who wants to go beyond using AI tools to understanding and building AI systems.
Your Pathway Through the Course
Essential Modules (Do These First)
| Order | Module | Why It Matters for You |
|---|---|---|
| 1 | M3 — LLMs & GenAI | Transformers, attention, tokenization — the architecture |
| 2 | M4 — Prompting | Advanced techniques: chaining, structured output, meta-prompting |
| 3 | M9 — Infrastructure | APIs, RAG pipelines, local models, cost engineering |
| 4 | M8 — Agents & GPTs | Building custom AI applications |
| 5 | M2 — How ML Works | Supervised, unsupervised, reinforcement — the foundations |
Recommended Additions
| Module | What You’ll Get |
|---|---|
| M10 — Ethics | Bias auditing, fairness metrics — essential for deployment |
| M7 — Multimodal | Vision models, diffusion, audio — expanding beyond text |
| M13 — Future | Where the field is heading — plan your skill development |
Key Resources
- Lab 9: RAG Pipeline Design — architecture a retrieval system
- Appendix: Tool Directory (Coding and Development section, Open Source section)
- Appendix: Prompt Library, Part 14 (Custom GPT Design)
- Appendix: Resources (fast.ai, Kaggle Learn, Hugging Face)
Your Capstone (Modified)
Build a working AI application: a custom GPT with uploaded knowledge files, a RAG-powered Q&A system, or an automation workflow connecting AI to external tools. Document: the problem it solves, architecture decisions, limitations, and a cost estimate for scaling.
Beyond This Course
This course provides the conceptual foundation. To go deeper technically:
- Python + APIs: fast.ai, Kaggle Learn, or Google’s ML Crash Course
- LLM Development: Hugging Face courses, Anthropic’s developer documentation
- Data Engineering: Google Data Analytics Certificate → dbt + SQL fundamentals
- Local Models: Start with Ollama, graduate to fine-tuning with QLoRA
Back to Modules