Skip to main content

Welcome to Tools in Data Science (TDS) — 2026

TDS is a hands-on course focused on modern data science workflows: developer tooling, LLM systems, agentic AI, security, and production MLOps.

Course roadmap​

The learning path is organized for progressive depth:

  • Weeks 1–2: Dev environment, version control, APIs, Docker, deployment
  • Weeks 3–4: Prompting, structured outputs, embeddings, RAG, retrieval quality
  • Weeks 5–8: Agentic systems, MCP, multimodal pipelines, CI/CD, security
  • Bonus Weeks 9–10: GCP MLOps, monitoring, scaling, and cost optimization

How to get the most from this course​

  1. Read the concept notes for context.
  2. Reproduce examples in your coding environment.
  3. Complete labs in sequence to build production intuition.
Recommended weekly flow

Spend one focused session on readings and one on implementation. This keeps conceptual understanding and practical execution in sync.

Using the integrated coding terminal​

You can open the terminal panel from any page:

  1. Press Ctrl+` (Control + backtick) or click the terminal button.
  2. (Linux, one-time) Install code-server:
bash
curl -fsSL https://code-server.dev/install.sh | sh
  1. Start your local code-server:
bash
code-server --auth none --bind-addr 127.0.0.1:8080
  1. In the terminal panel choose:
    • Localhost → port 8080
    • GitHub Codespaces → open/manage Codespaces inside the panel
    • Custom URL → paste any http:// or https:// URL (including forwarded ports)
Save your work

Always push your work to GitHub frequently. See Week 1 Git & GitHub.

Prerequisites​

  • Basic Python programming
  • Comfort with the command line
  • Intro-level SQL
  • Willingness to build and debug real projects

No prior Docker, LLM, or cloud deployment experience is required.

Start here​

Go to Week 1 — Dev Environment & Version Control and complete it before moving on.