On-ramp course blending AI-infrastructure literacy and developer tooling; does not replace CIS programming, data-science modeling, cloud administration, or cybersecurity fundamentals. Deep containers/Kubernetes/cloud are covered in AIINFRA 101 and deep security in AIINFRA 302.
← All courses
// AIINFRA 100 · Semester 1
AI Systems & Development Environments
The on-ramp: AI systems literacy plus a professional Python/Git/ML dev environment
Combines AI-infrastructure literacy with the hands-on developer toolkit. Students learn what AI workloads are and the compute, data, and software layers they depend on, then set up a professional Python, VS Code, and Git environment, manage reproducible environments, and use the core data-science and ML toolchain (NumPy, pandas, scikit-learn, and an introduction to PyTorch). The first course in the certificate; no prior AI experience required. Deep container, Kubernetes, and cloud work is covered in AIINFRA 101, and deep security in AIINFRA 302.
download / canvas package
Canvas IMSCC Export
Canvas IMSCC Export
Download the complete Canvas import package for AIINFRA 100.
01 / outcomes
Outcomes
Course objectives
- Define key AI-infrastructure concepts and identify appropriate compute, data, and software components for a given AI workload.
- Compare cloud, on-premises, and hybrid deployment models against real requirements, cost, and constraints.
- Configure and customize a professional AI development environment (VS Code, Python 3.11+, extensions).
- Use Git and GitHub for version control and collaboration, and manage reproducible environments with venv, conda, and uv.
- Use the core Python data-science/ML toolchain (NumPy, pandas, Jupyter, scikit-learn, intro PyTorch) and track experiments with MLflow.
Student learning outcomes
- Identify infrastructure components (compute/data/software) for a given AI workload
- Evaluate cloud vs. on-prem vs. hybrid deployment models against requirements and cost
- Configure a professional AI dev environment (VS Code, Python, extensions)
- Use Git/GitHub and reproducible environments (venv, conda, uv)
- Use the core data-science/ML toolchain (NumPy, pandas, scikit-learn, PyTorch) and MLflow
02 / schedule
16-week schedule
Wk 01
Introduction to AI Infrastructure
Frames the prototype-vs-production theme and the layered AI-infrastructure mental model.
Wk 02
AI Workloads: Training vs. Inference
Distinguishes training and inference workloads and the latency-vs-throughput trade-offs each demands.
Wk 03
Computational Resources: CPUs, GPUs, TPUs & VRAM
Surveys AI accelerators and why VRAM, not just speed, is the real bottleneck.
Wk 04
The AI Software Stack
Frameworks, CUDA, drivers, containers, and model formats that underlie AI systems.
Wk 05
Python for AI — Fundamentals
Core Python for AI work: data structures, functions, and running Python like you mean it.
Wk 06
The Developer Toolbox & VS Code
Turns VS Code into an AI/ML workstation with the right extensions and workflow.
Wk 07
Version Control with Git
Repos, snapshots, branching, and keeping data out of history.
Wk 08
Collaborative Development with GitHub — Midterm
Remotes, pull requests, and the collaboration flow; includes the course midterm.
Midterm · covers Wks 1–7Wk 09
Environment & Package Management
Isolating and reproducing Python setups with venv, conda, and modern uv.
Wk 10
Jupyter & the Data-Science Stack
The Jupyter ecosystem plus NumPy arrays and pandas DataFrames.
Wk 11
Machine Learning Libraries: scikit-learn
The estimator API and the train/predict/evaluate workflow.
Wk 12
Deep Learning Tooling: PyTorch
Tensors, autograd, and a first neural network.
Wk 13
Data & Storage Infrastructure for AI
Warehouses, lakes, formats, and why file format is an AI decision.
Wk 14
Cloud, On-Prem & Hybrid Deployment
Deployment models and cost awareness across cloud, on-prem, and hybrid, with FinOps basics.
Wk 15
Reproducibility, Testing & Responsible AI
Testing/QA, experiment tracking with MLflow, and responsible-AI governance (NIST AI RMF, EU AI Act).
Wk 16
Final Project & Course Review
Students complete a final portfolio project and review the course as a whole.
Capstone03 / Senate-ready controls
Course evidence and boundaries
These controls make the curriculum reviewable without changing the 10-course sequence.
AI workload infrastructure decision memo plus a professional, version-controlled AI dev environment.
Survey-level OWASP LLM Top 10 and responsible-AI (NIST AI RMF / EU AI Act) risk note.
Reproducible project scaffold (env + Git + notebook/ML toolchain), infrastructure plan, and short presentation.
04 / tools
Tools & frameworks
Local Development
Environments & Packaging
Data Science & ML
Cloud Platforms
Reproducibility & Governance